Multi-model analysis in hydrological prediction

Science.gov (United States)

Lanthier, M.; Arsenault, R.; Brissette, F.

2017-12-01

Hydrologic modelling, by nature, is a simplification of the real-world hydrologic system. Therefore ensemble hydrological predictions thus obtained do not present the full range of possible streamflow outcomes, thereby producing ensembles which demonstrate errors in variance such as under-dispersion. Past studies show that lumped models used in prediction mode can return satisfactory results, especially when there is not enough information available on the watershed to run a distributed model. But all lumped models greatly simplify the complex processes of the hydrologic cycle. To generate more spread in the hydrologic ensemble predictions, multi-model ensembles have been considered. In this study, the aim is to propose and analyse a method that gives an ensemble streamflow prediction that properly represents the forecast probabilities and reduced ensemble bias. To achieve this, three simple lumped models are used to generate an ensemble. These will also be combined using multi-model averaging techniques, which generally generate a more accurate hydrogram than the best of the individual models in simulation mode. This new predictive combined hydrogram is added to the ensemble, thus creating a large ensemble which may improve the variability while also improving the ensemble mean bias. The quality of the predictions is then assessed on different periods: 2 weeks, 1 month, 3 months and 6 months using a PIT Histogram of the percentiles of the real observation volumes with respect to the volumes of the ensemble members. Initially, the models were run using historical weather data to generate synthetic flows. This worked for individual models, but not for the multi-model and for the large ensemble. Consequently, by performing data assimilation at each prediction period and thus adjusting the initial states of the models, the PIT Histogram could be constructed using the observed flows while allowing the use of the multi-model predictions. The under-dispersion has been

Integrating SMOS brightness temperatures with a new conceptual spatially distributed hydrological model for improving flood and drought predictions at large scale.

Science.gov (United States)

Hostache, Renaud; Rains, Dominik; Chini, Marco; Lievens, Hans; Verhoest, Niko E. C.; Matgen, Patrick

2017-04-01

Motivated by climate change and its impact on the scarcity or excess of water in many parts of the world, several agencies and research institutions have taken initiatives in monitoring and predicting the hydrologic cycle at a global scale. Such a monitoring/prediction effort is important for understanding the vulnerability to extreme hydrological events and for providing early warnings. This can be based on an optimal combination of hydro-meteorological models and remote sensing, in which satellite measurements can be used as forcing or calibration data or for regularly updating the model states or parameters. Many advances have been made in these domains and the near future will bring new opportunities with respect to remote sensing as a result of the increasing number of spaceborn sensors enabling the large scale monitoring of water resources. Besides of these advances, there is currently a tendency to refine and further complicate physically-based hydrologic models to better capture the hydrologic processes at hand. However, this may not necessarily be beneficial for large-scale hydrology, as computational efforts are therefore increasing significantly. As a matter of fact, a novel thematic science question that is to be investigated is whether a flexible conceptual model can match the performance of a complex physically-based model for hydrologic simulations at large scale. In this context, the main objective of this study is to investigate how innovative techniques that allow for the estimation of soil moisture from satellite data can help in reducing errors and uncertainties in large scale conceptual hydro-meteorological modelling. A spatially distributed conceptual hydrologic model has been set up based on recent developments of the SUPERFLEX modelling framework. As it requires limited computational efforts, this model enables early warnings for large areas. Using as forcings the ERA-Interim public dataset and coupled with the CMEM radiative transfer model

Spatiotemporal property and predictability of large-scale human mobility

Science.gov (United States)

Zhang, Hai-Tao; Zhu, Tao; Fu, Dongfei; Xu, Bowen; Han, Xiao-Pu; Chen, Duxin

2018-04-01

Spatiotemporal characteristics of human mobility emerging from complexity on individual scale have been extensively studied due to the application potential on human behavior prediction and recommendation, and control of epidemic spreading. We collect and investigate a comprehensive data set of human activities on large geographical scales, including both websites browse and mobile towers visit. Numerical results show that the degree of activity decays as a power law, indicating that human behaviors are reminiscent of scale-free random walks known as Lévy flight. More significantly, this study suggests that human activities on large geographical scales have specific non-Markovian characteristics, such as a two-segment power-law distribution of dwelling time and a high possibility for prediction. Furthermore, a scale-free featured mobility model with two essential ingredients, i.e., preferential return and exploration, and a Gaussian distribution assumption on the exploration tendency parameter is proposed, which outperforms existing human mobility models under scenarios of large geographical scales.

Large-area dry bean yield prediction modeling in Mexico

Science.gov (United States)

Given the importance of dry bean in Mexico, crop yield predictions before harvest are valuable for authorities of the agricultural sector, in order to define support for producers. The aim of this study was to develop an empirical model to estimate the yield of dry bean at the regional level prior t...

Prediction Models for Dynamic Demand Response

Energy Technology Data Exchange (ETDEWEB)

Aman, Saima; Frincu, Marc; Chelmis, Charalampos; Noor, Muhammad; Simmhan, Yogesh; Prasanna, Viktor K.

2015-11-02

As Smart Grids move closer to dynamic curtailment programs, Demand Response (DR) events will become necessary not only on fixed time intervals and weekdays predetermined by static policies, but also during changing decision periods and weekends to react to real-time demand signals. Unique challenges arise in this context vis-a-vis demand prediction and curtailment estimation and the transformation of such tasks into an automated, efficient dynamic demand response (D²R) process. While existing work has concentrated on increasing the accuracy of prediction models for DR, there is a lack of studies for prediction models for D²R, which we address in this paper. Our first contribution is the formal definition of D²R, and the description of its challenges and requirements. Our second contribution is a feasibility analysis of very-short-term prediction of electricity consumption for D²R over a diverse, large-scale dataset that includes both small residential customers and large buildings. Our third, and major contribution is a set of insights into the predictability of electricity consumption in the context of D²R. Specifically, we focus on prediction models that can operate at a very small data granularity (here 15-min intervals), for both weekdays and weekends - all conditions that characterize scenarios for D²R. We find that short-term time series and simple averaging models used by Independent Service Operators and utilities achieve superior prediction accuracy. We also observe that workdays are more predictable than weekends and holiday. Also, smaller customers have large variation in consumption and are less predictable than larger buildings. Key implications of our findings are that better models are required for small customers and for non-workdays, both of which are critical for D²R. Also, prediction models require just few days’ worth of data indicating that small amounts of

Predicting climate-induced range shifts: model differences and model reliability.

Science.gov (United States)

Joshua J. Lawler; Denis White; Ronald P. Neilson; Andrew R. Blaustein

2006-01-01

Predicted changes in the global climate are likely to cause large shifts in the geographic ranges of many plant and animal species. To date, predictions of future range shifts have relied on a variety of modeling approaches with different levels of model accuracy. Using a common data set, we investigated the potential implications of alternative modeling approaches for...

Quantitative Missense Variant Effect Prediction Using Large-Scale Mutagenesis Data.

Science.gov (United States)

Gray, Vanessa E; Hause, Ronald J; Luebeck, Jens; Shendure, Jay; Fowler, Douglas M

2018-01-24

Large datasets describing the quantitative effects of mutations on protein function are becoming increasingly available. Here, we leverage these datasets to develop Envision, which predicts the magnitude of a missense variant's molecular effect. Envision combines 21,026 variant effect measurements from nine large-scale experimental mutagenesis datasets, a hitherto untapped training resource, with a supervised, stochastic gradient boosting learning algorithm. Envision outperforms other missense variant effect predictors both on large-scale mutagenesis data and on an independent test dataset comprising 2,312 TP53 variants whose effects were measured using a low-throughput approach. This dataset was never used for hyperparameter tuning or model training and thus serves as an independent validation set. Envision prediction accuracy is also more consistent across amino acids than other predictors. Finally, we demonstrate that Envision's performance improves as more large-scale mutagenesis data are incorporated. We precompute Envision predictions for every possible single amino acid variant in human, mouse, frog, zebrafish, fruit fly, worm, and yeast proteomes (https://envision.gs.washington.edu/). Copyright © 2017 Elsevier Inc. All rights reserved.

Distributed Model Predictive Control over Multiple Groups of Vehicles in Highway Intelligent Space for Large Scale System

Directory of Open Access Journals (Sweden)

Tang Xiaofeng

2014-01-01

Full Text Available The paper presents the three time warning distances for solving the large scale system of multiple groups of vehicles safety driving characteristics towards highway tunnel environment based on distributed model prediction control approach. Generally speaking, the system includes two parts. First, multiple vehicles are divided into multiple groups. Meanwhile, the distributed model predictive control approach is proposed to calculate the information framework of each group. Each group of optimization performance considers the local optimization and the neighboring subgroup of optimization characteristics, which could ensure the global optimization performance. Second, the three time warning distances are studied based on the basic principles used for highway intelligent space (HIS and the information framework concept is proposed according to the multiple groups of vehicles. The math model is built to avoid the chain avoidance of vehicles. The results demonstrate that the proposed highway intelligent space method could effectively ensure driving safety of multiple groups of vehicles under the environment of fog, rain, or snow.

Genomic prediction of complex human traits: relatedness, trait architecture and predictive meta-models

Science.gov (United States)

Spiliopoulou, Athina; Nagy, Reka; Bermingham, Mairead L.; Huffman, Jennifer E.; Hayward, Caroline; Vitart, Veronique; Rudan, Igor; Campbell, Harry; Wright, Alan F.; Wilson, James F.; Pong-Wong, Ricardo; Agakov, Felix; Navarro, Pau; Haley, Chris S.

2015-01-01

We explore the prediction of individuals' phenotypes for complex traits using genomic data. We compare several widely used prediction models, including Ridge Regression, LASSO and Elastic Nets estimated from cohort data, and polygenic risk scores constructed using published summary statistics from genome-wide association meta-analyses (GWAMA). We evaluate the interplay between relatedness, trait architecture and optimal marker density, by predicting height, body mass index (BMI) and high-density lipoprotein level (HDL) in two data cohorts, originating from Croatia and Scotland. We empirically demonstrate that dense models are better when all genetic effects are small (height and BMI) and target individuals are related to the training samples, while sparse models predict better in unrelated individuals and when some effects have moderate size (HDL). For HDL sparse models achieved good across-cohort prediction, performing similarly to the GWAMA risk score and to models trained within the same cohort, which indicates that, for predicting traits with moderately sized effects, large sample sizes and familial structure become less important, though still potentially useful. Finally, we propose a novel ensemble of whole-genome predictors with GWAMA risk scores and demonstrate that the resulting meta-model achieves higher prediction accuracy than either model on its own. We conclude that although current genomic predictors are not accurate enough for diagnostic purposes, performance can be improved without requiring access to large-scale individual-level data. Our methodologically simple meta-model is a means of performing predictive meta-analysis for optimizing genomic predictions and can be easily extended to incorporate multiple population-level summary statistics or other domain knowledge. PMID:25918167

Large Hadron Collider (LHC) phenomenology, operational challenges and theoretical predictions

CERN Document Server

Gilles, Abelin R

2013-01-01

The Large Hadron Collider (LHC) is the highest-energy particle collider ever constructed and is considered "one of the great engineering milestones of mankind." It was built by the European Organization for Nuclear Research (CERN) from 1998 to 2008, with the aim of allowing physicists to test the predictions of different theories of particle physics and high-energy physics, and particularly prove or disprove the existence of the theorized Higgs boson and of the large family of new particles predicted by supersymmetric theories. In this book, the authors study the phenomenology, operational challenges and theoretical predictions of LHC. Topics discussed include neutral and charged black hole remnants at the LHC; the modified statistics approach for the thermodynamical model of multiparticle production; and astroparticle physics and cosmology in the LHC era.

The research of the quantitative prediction of the deposits concentrated regions of the large and super-large sized mineral deposits in China

International Nuclear Information System (INIS)

Zhao Zhenyu; Wang Shicheng

2003-01-01

By the general theory and method of mineral resources prognosis of synthetic information, the locative and quantitative prediction of the large and super-large sized mineral deposits of solid resources of 1 : 5,000,000 are developed in china. The deposit concentrated regions is model unit, the anomaly concentrated regions is prediction unit. The mineral prognosis of synthetic information is developed on GIS platform. The technical route and work method of looking for the large and super-large sized mineral resources and basic principle of compiling attribute table of the variables and the response variables are mentioned. In research of prediction of resources quantity, the locative and quantitative prediction are processed by separately the quantification theory Ⅲ and the corresponding characteristic analysis, two methods are compared. It is very important for resources prediction of western ten provinces in china, it is helpful. (authors)

Large-scale transportation network congestion evolution prediction using deep learning theory.

Science.gov (United States)

Ma, Xiaolei; Yu, Haiyang; Wang, Yunpeng; Wang, Yinhai

2015-01-01

Understanding how congestion at one location can cause ripples throughout large-scale transportation network is vital for transportation researchers and practitioners to pinpoint traffic bottlenecks for congestion mitigation. Traditional studies rely on either mathematical equations or simulation techniques to model traffic congestion dynamics. However, most of the approaches have limitations, largely due to unrealistic assumptions and cumbersome parameter calibration process. With the development of Intelligent Transportation Systems (ITS) and Internet of Things (IoT), transportation data become more and more ubiquitous. This triggers a series of data-driven research to investigate transportation phenomena. Among them, deep learning theory is considered one of the most promising techniques to tackle tremendous high-dimensional data. This study attempts to extend deep learning theory into large-scale transportation network analysis. A deep Restricted Boltzmann Machine and Recurrent Neural Network architecture is utilized to model and predict traffic congestion evolution based on Global Positioning System (GPS) data from taxi. A numerical study in Ningbo, China is conducted to validate the effectiveness and efficiency of the proposed method. Results show that the prediction accuracy can achieve as high as 88% within less than 6 minutes when the model is implemented in a Graphic Processing Unit (GPU)-based parallel computing environment. The predicted congestion evolution patterns can be visualized temporally and spatially through a map-based platform to identify the vulnerable links for proactive congestion mitigation.

Ensemble modeling to predict habitat suitability for a large-scale disturbance specialist

Science.gov (United States)

Quresh S. Latif; Victoria A. Saab; Jonathan G. Dudley; Jeff P. Hollenbeck

2013-01-01

To conserve habitat for disturbance specialist species, ecologists must identify where individuals will likely settle in newly disturbed areas. Habitat suitability models can predict which sites at new disturbances will most likely attract specialists. Without validation data from newly disturbed areas, however, the best approach for maximizing predictive accuracy can...

Large-Scale Mapping and Predictive Modeling of Submerged Aquatic Vegetation in a Shallow Eutrophic Lake

Directory of Open Access Journals (Sweden)

Karl E. Havens

2002-01-01

Full Text Available A spatially intensive sampling program was developed for mapping the submerged aquatic vegetation (SAV over an area of approximately 20,000 ha in a large, shallow lake in Florida, U.S. The sampling program integrates Geographic Information System (GIS technology with traditional field sampling of SAV and has the capability of producing robust vegetation maps under a wide range of conditions, including high turbidity, variable depth (0 to 2 m, and variable sediment types. Based on sampling carried out in AugustœSeptember 2000, we measured 1,050 to 4,300 ha of vascular SAV species and approximately 14,000 ha of the macroalga Chara spp. The results were similar to those reported in the early 1990s, when the last large-scale SAV sampling occurred. Occurrence of Chara was strongly associated with peat sediments, and maximal depths of occurrence varied between sediment types (mud, sand, rock, and peat. A simple model of Chara occurrence, based only on water depth, had an accuracy of 55%. It predicted occurrence of Chara over large areas where the plant actually was not found. A model based on sediment type and depth had an accuracy of 75% and produced a spatial map very similar to that based on observations. While this approach needs to be validated with independent data in order to test its general utility, we believe it may have application elsewhere. The simple modeling approach could serve as a coarse-scale tool for evaluating effects of water level management on Chara populations.

Large-scale mapping and predictive modeling of submerged aquatic vegetation in a shallow eutrophic lake.

Science.gov (United States)

Havens, Karl E; Harwell, Matthew C; Brady, Mark A; Sharfstein, Bruce; East, Therese L; Rodusky, Andrew J; Anson, Daniel; Maki, Ryan P

2002-04-09

A spatially intensive sampling program was developed for mapping the submerged aquatic vegetation (SAV) over an area of approximately 20,000 ha in a large, shallow lake in Florida, U.S. The sampling program integrates Geographic Information System (GIS) technology with traditional field sampling of SAV and has the capability of producing robust vegetation maps under a wide range of conditions, including high turbidity, variable depth (0 to 2 m), and variable sediment types. Based on sampling carried out in August-September 2000, we measured 1,050 to 4,300 ha of vascular SAV species and approximately 14,000 ha of the macroalga Chara spp. The results were similar to those reported in the early 1990s, when the last large-scale SAV sampling occurred. Occurrence of Chara was strongly associated with peat sediments, and maximal depths of occurrence varied between sediment types (mud, sand, rock, and peat). A simple model of Chara occurrence, based only on water depth, had an accuracy of 55%. It predicted occurrence of Chara over large areas where the plant actually was not found. A model based on sediment type and depth had an accuracy of 75% and produced a spatial map very similar to that based on observations. While this approach needs to be validated with independent data in order to test its general utility, we believe it may have application elsewhere. The simple modeling approach could serve as a coarse-scale tool for evaluating effects of water level management on Chara populations.

Dynamic subgrid scale model used in a deep bundle turbulence prediction using the large eddy simulation method

International Nuclear Information System (INIS)

Barsamian, H.R.; Hassan, Y.A.

1996-01-01

Turbulence is one of the most commonly occurring phenomena of engineering interest in the field of fluid mechanics. Since most flows are turbulent, there is a significant payoff for improved predictive models of turbulence. One area of concern is the turbulent buffeting forces experienced by the tubes in steam generators of nuclear power plants. Although the Navier-Stokes equations are able to describe turbulent flow fields, the large number of scales of turbulence limit practical flow field calculations with current computing power. The dynamic subgrid scale closure model of Germano et. al (1991) is used in the large eddy simulation code GUST for incompressible isothermal flows. Tube bundle geometries of staggered and non-staggered arrays are considered in deep bundle simulations. The advantage of the dynamic subgrid scale model is the exclusion of an input model coefficient. The model coefficient is evaluated dynamically for each nodal location in the flow domain. Dynamic subgrid scale results are obtained in the form of power spectral densities and flow visualization of turbulent characteristics. Comparisons are performed among the dynamic subgrid scale model, the Smagorinsky eddy viscosity model (Smagorinsky, 1963) (that is used as the base model for the dynamic subgrid scale model) and available experimental data. Spectral results of the dynamic subgrid scale model correlate better with experimental data. Satisfactory turbulence characteristics are observed through flow visualization

Prediction of Thermal Environment in a Large Space Using Artificial Neural Network

Directory of Open Access Journals (Sweden)

Hyun-Jung Yoon

2018-02-01

Full Text Available Since the thermal environment of large space buildings such as stadiums can vary depending on the location of the stands, it is important to divide them into different zones and evaluate their thermal environment separately. The thermal environment can be evaluated using physical values measured with the sensors, but the occupant density of the stadium stands is high, which limits the locations available to install the sensors. As a method to resolve the limitations of installing the sensors, we propose a method to predict the thermal environment of each zone in a large space. We set six key thermal factors affecting the thermal environment in a large space to be predicted factors (indoor air temperature, mean radiant temperature, and clothing and the fixed factors (air velocity, metabolic rate, and relative humidity. Using artificial neural network (ANN models and the outdoor air temperature and the surface temperature of the interior walls around the stands as input data, we developed a method to predict the three thermal factors. Learning and verification datasets were established using STAR CCM+ (2016.10, Siemens PLM software, Plano, TX, USA. An analysis of each model’s prediction results showed that the prediction accuracy increased with the number of learning data points. The thermal environment evaluation process developed in this study can be used to control heating, ventilation, and air conditioning (HVAC facilities in each zone in a large space building with sufficient learning by ANN models at the building testing or the evaluation stage.

Spatial Economics Model Predicting Transport Volume

Directory of Open Access Journals (Sweden)

Lu Bo

2016-10-01

Full Text Available It is extremely important to predict the logistics requirements in a scientific and rational way. However, in recent years, the improvement effect on the prediction method is not very significant and the traditional statistical prediction method has the defects of low precision and poor interpretation of the prediction model, which cannot only guarantee the generalization ability of the prediction model theoretically, but also cannot explain the models effectively. Therefore, in combination with the theories of the spatial economics, industrial economics, and neo-classical economics, taking city of Zhuanghe as the research object, the study identifies the leading industry that can produce a large number of cargoes, and further predicts the static logistics generation of the Zhuanghe and hinterlands. By integrating various factors that can affect the regional logistics requirements, this study established a logistics requirements potential model from the aspect of spatial economic principles, and expanded the way of logistics requirements prediction from the single statistical principles to an new area of special and regional economics.

Large-scale transportation network congestion evolution prediction using deep learning theory.

Directory of Open Access Journals (Sweden)

Xiaolei Ma

Full Text Available Understanding how congestion at one location can cause ripples throughout large-scale transportation network is vital for transportation researchers and practitioners to pinpoint traffic bottlenecks for congestion mitigation. Traditional studies rely on either mathematical equations or simulation techniques to model traffic congestion dynamics. However, most of the approaches have limitations, largely due to unrealistic assumptions and cumbersome parameter calibration process. With the development of Intelligent Transportation Systems (ITS and Internet of Things (IoT, transportation data become more and more ubiquitous. This triggers a series of data-driven research to investigate transportation phenomena. Among them, deep learning theory is considered one of the most promising techniques to tackle tremendous high-dimensional data. This study attempts to extend deep learning theory into large-scale transportation network analysis. A deep Restricted Boltzmann Machine and Recurrent Neural Network architecture is utilized to model and predict traffic congestion evolution based on Global Positioning System (GPS data from taxi. A numerical study in Ningbo, China is conducted to validate the effectiveness and efficiency of the proposed method. Results show that the prediction accuracy can achieve as high as 88% within less than 6 minutes when the model is implemented in a Graphic Processing Unit (GPU-based parallel computing environment. The predicted congestion evolution patterns can be visualized temporally and spatially through a map-based platform to identify the vulnerable links for proactive congestion mitigation.

A hydrogeomorphic river network model predicts where and why hyporheic exchange is important in large basins

Science.gov (United States)

Gomez-Velez, Jesus D.; Harvey, Judson W.

2014-09-01

Hyporheic exchange has been hypothesized to have basin-scale consequences; however, predictions throughout river networks are limited by available geomorphic and hydrogeologic data and by models that can analyze and aggregate hyporheic exchange flows across large spatial scales. We developed a parsimonious but physically based model of hyporheic flow for application in large river basins: Networks with EXchange and Subsurface Storage (NEXSS). We applied NEXSS across a broad range of geomorphic diversity in river reaches and synthetic river networks. NEXSS demonstrates that vertical exchange beneath submerged bed forms rather than lateral exchange through meanders dominates hyporheic fluxes and turnover rates along river corridors. Per kilometer, low-order streams have a biogeochemical potential at least 2 orders of magnitude larger than higher-order streams. However, when biogeochemical potential is examined per average length of each stream order, low- and high-order streams were often found to be comparable. As a result, the hyporheic zone's intrinsic potential for biogeochemical transformations is comparable across different stream orders, but the greater river miles and larger total streambed area of lower order streams result in the highest cumulative impact from low-order streams. Lateral exchange through meander banks may be important in some cases but generally only in large rivers.

A hydrogeomorphic river network model predicts where and why hyporheic exchange is important in large basins

Science.gov (United States)

Gomez-Velez, Jesus D.; Harvey, Judson

2014-01-01

Hyporheic exchange has been hypothesized to have basin-scale consequences; however, predictions throughout river networks are limited by available geomorphic and hydrogeologic data and by models that can analyze and aggregate hyporheic exchange flows across large spatial scales. We developed a parsimonious but physically based model of hyporheic flow for application in large river basins: Networks with EXchange and Subsurface Storage (NEXSS). We applied NEXSS across a broad range of geomorphic diversity in river reaches and synthetic river networks. NEXSS demonstrates that vertical exchange beneath submerged bed forms rather than lateral exchange through meanders dominates hyporheic fluxes and turnover rates along river corridors. Per kilometer, low-order streams have a biogeochemical potential at least 2 orders of magnitude larger than higher-order streams. However, when biogeochemical potential is examined per average length of each stream order, low- and high-order streams were often found to be comparable. As a result, the hyporheic zone's intrinsic potential for biogeochemical transformations is comparable across different stream orders, but the greater river miles and larger total streambed area of lower order streams result in the highest cumulative impact from low-order streams. Lateral exchange through meander banks may be important in some cases but generally only in large rivers.

Predicting Positive and Negative Relationships in Large Social Networks.

Directory of Open Access Journals (Sweden)

Guan-Nan Wang

Full Text Available In a social network, users hold and express positive and negative attitudes (e.g. support/opposition towards other users. Those attitudes exhibit some kind of binary relationships among the users, which play an important role in social network analysis. However, some of those binary relationships are likely to be latent as the scale of social network increases. The essence of predicting latent binary relationships have recently began to draw researchers' attention. In this paper, we propose a machine learning algorithm for predicting positive and negative relationships in social networks inspired by structural balance theory and social status theory. More specifically, we show that when two users in the network have fewer common neighbors, the prediction accuracy of the relationship between them deteriorates. Accordingly, in the training phase, we propose a segment-based training framework to divide the training data into two subsets according to the number of common neighbors between users, and build a prediction model for each subset based on support vector machine (SVM. Moreover, to deal with large-scale social network data, we employ a sampling strategy that selects small amount of training data while maintaining high accuracy of prediction. We compare our algorithm with traditional algorithms and adaptive boosting of them. Experimental results of typical data sets show that our algorithm can deal with large social networks and consistently outperforms other methods.

Predicting Positive and Negative Relationships in Large Social Networks.

Science.gov (United States)

Wang, Guan-Nan; Gao, Hui; Chen, Lian; Mensah, Dennis N A; Fu, Yan

2015-01-01

In a social network, users hold and express positive and negative attitudes (e.g. support/opposition) towards other users. Those attitudes exhibit some kind of binary relationships among the users, which play an important role in social network analysis. However, some of those binary relationships are likely to be latent as the scale of social network increases. The essence of predicting latent binary relationships have recently began to draw researchers' attention. In this paper, we propose a machine learning algorithm for predicting positive and negative relationships in social networks inspired by structural balance theory and social status theory. More specifically, we show that when two users in the network have fewer common neighbors, the prediction accuracy of the relationship between them deteriorates. Accordingly, in the training phase, we propose a segment-based training framework to divide the training data into two subsets according to the number of common neighbors between users, and build a prediction model for each subset based on support vector machine (SVM). Moreover, to deal with large-scale social network data, we employ a sampling strategy that selects small amount of training data while maintaining high accuracy of prediction. We compare our algorithm with traditional algorithms and adaptive boosting of them. Experimental results of typical data sets show that our algorithm can deal with large social networks and consistently outperforms other methods.

Constituent models and large transverse momentum reactions

International Nuclear Information System (INIS)

Brodsky, S.J.

1975-01-01

The discussion of constituent models and large transverse momentum reactions includes the structure of hard scattering models, dimensional counting rules for large transverse momentum reactions, dimensional counting and exclusive processes, the deuteron form factor, applications to inclusive reactions, predictions for meson and photon beams, the charge-cubed test for the e/sup +-/p → e/sup +-/γX asymmetry, the quasi-elastic peak in inclusive hadronic reactions, correlations, and the multiplicity bump at large transverse momentum. Also covered are the partition method for bound state calculations, proofs of dimensional counting, minimal neutralization and quark--quark scattering, the development of the constituent interchange model, and the A dependence of high transverse momentum reactions

Modelling and control of large cryogenic refrigerator

International Nuclear Information System (INIS)

Bonne, Francois

2014-01-01

This manuscript is concern with both the modeling and the derivation of control schemes for large cryogenic refrigerators. The particular case of those which are submitted to highly variable pulsed heat load is studied. A model of each object that normally compose a large cryo-refrigerator is proposed. The methodology to gather objects model into the model of a subsystem is presented. The manuscript also shows how to obtain a linear equivalent model of the subsystem. Based on the derived models, advances control scheme are proposed. Precisely, a linear quadratic controller for warm compression station working with both two and three pressures state is derived, and a predictive constrained one for the cold-box is obtained. The particularity of those control schemes is that they fit the computing and data storage capabilities of Programmable Logic Controllers (PLC) with are well used in industry. The open loop model prediction capability is assessed using experimental data. Developed control schemes are validated in simulation and experimentally on the 400W1.8K SBT's cryogenic test facility and on the CERN's LHC warm compression station. (author) [fr

Evaluation of burst pressure prediction models for line pipes

Energy Technology Data Exchange (ETDEWEB)

Zhu, Xian-Kui, E-mail: zhux@battelle.org [Battelle Memorial Institute, 505 King Avenue, Columbus, OH 43201 (United States); Leis, Brian N. [Battelle Memorial Institute, 505 King Avenue, Columbus, OH 43201 (United States)

2012-01-15

Accurate prediction of burst pressure plays a central role in engineering design and integrity assessment of oil and gas pipelines. Theoretical and empirical solutions for such prediction are evaluated in this paper relative to a burst pressure database comprising more than 100 tests covering a variety of pipeline steel grades and pipe sizes. Solutions considered include three based on plasticity theory for the end-capped, thin-walled, defect-free line pipe subjected to internal pressure in terms of the Tresca, von Mises, and ZL (or Zhu-Leis) criteria, one based on a cylindrical instability stress (CIS) concept, and a large group of analytical and empirical models previously evaluated by Law and Bowie (International Journal of Pressure Vessels and Piping, 84, 2007: 487-492). It is found that these models can be categorized into either a Tresca-family or a von Mises-family of solutions, except for those due to Margetson and Zhu-Leis models. The viability of predictions is measured via statistical analyses in terms of a mean error and its standard deviation. Consistent with an independent parallel evaluation using another large database, the Zhu-Leis solution is found best for predicting burst pressure, including consideration of strain hardening effects, while the Tresca strength solutions including Barlow, Maximum shear stress, Turner, and the ASME boiler code provide reasonably good predictions for the class of line-pipe steels with intermediate strain hardening response. - Highlights: Black-Right-Pointing-Pointer This paper evaluates different burst pressure prediction models for line pipes. Black-Right-Pointing-Pointer The existing models are categorized into two major groups of Tresca and von Mises solutions. Black-Right-Pointing-Pointer Prediction quality of each model is assessed statistically using a large full-scale burst test database. Black-Right-Pointing-Pointer The Zhu-Leis solution is identified as the best predictive model.

Evaluation of burst pressure prediction models for line pipes

International Nuclear Information System (INIS)

Zhu, Xian-Kui; Leis, Brian N.

2012-01-01

Accurate prediction of burst pressure plays a central role in engineering design and integrity assessment of oil and gas pipelines. Theoretical and empirical solutions for such prediction are evaluated in this paper relative to a burst pressure database comprising more than 100 tests covering a variety of pipeline steel grades and pipe sizes. Solutions considered include three based on plasticity theory for the end-capped, thin-walled, defect-free line pipe subjected to internal pressure in terms of the Tresca, von Mises, and ZL (or Zhu-Leis) criteria, one based on a cylindrical instability stress (CIS) concept, and a large group of analytical and empirical models previously evaluated by Law and Bowie (International Journal of Pressure Vessels and Piping, 84, 2007: 487–492). It is found that these models can be categorized into either a Tresca-family or a von Mises-family of solutions, except for those due to Margetson and Zhu-Leis models. The viability of predictions is measured via statistical analyses in terms of a mean error and its standard deviation. Consistent with an independent parallel evaluation using another large database, the Zhu-Leis solution is found best for predicting burst pressure, including consideration of strain hardening effects, while the Tresca strength solutions including Barlow, Maximum shear stress, Turner, and the ASME boiler code provide reasonably good predictions for the class of line-pipe steels with intermediate strain hardening response. - Highlights: ► This paper evaluates different burst pressure prediction models for line pipes. ► The existing models are categorized into two major groups of Tresca and von Mises solutions. ► Prediction quality of each model is assessed statistically using a large full-scale burst test database. ► The Zhu-Leis solution is identified as the best predictive model.

Predicting the performance of a power amplifier using large-signal circuit simulations of an AlGaN/GaN HFET model

Science.gov (United States)

Bilbro, Griff L.; Hou, Danqiong; Yin, Hong; Trew, Robert J.

2009-02-01

We have quantitatively modeled the conduction current and charge storage of an HFET in terms its physical dimensions and material properties. For DC or small-signal RF operation, no adjustable parameters are necessary to predict the terminal characteristics of the device. Linear performance measures such as small-signal gain and input admittance can be predicted directly from the geometric structure and material properties assumed for the device design. We have validated our model at low-frequency against experimental I-V measurements and against two-dimensional device simulations. We discuss our recent extension of our model to include a larger class of electron velocity-field curves. We also discuss the recent reformulation of our model to facilitate its implementation in commercial large-signal high-frequency circuit simulators. Large signal RF operation is more complex. First, the highest CW microwave power is fundamentally bounded by a brief, reversible channel breakdown in each RF cycle. Second, the highest experimental measurements of efficiency, power, or linearity always require harmonic load pull and possibly also harmonic source pull. Presently, our model accounts for these facts with an adjustable breakdown voltage and with adjustable load impedances and source impedances for the fundamental frequency and its harmonics. This has allowed us to validate our model for large signal RF conditions by simultaneously fitting experimental measurements of output power, gain, and power added efficiency of real devices. We show that the resulting model can be used to compare alternative device designs in terms of their large signal performance, such as their output power at 1dB gain compression or their third order intercept points. In addition, the model provides insight into new device physics features enabled by the unprecedented current and voltage levels of AlGaN/GaN HFETs, including non-ohmic resistance in the source access regions and partial depletion of

Neural Fuzzy Inference System-Based Weather Prediction Model and Its Precipitation Predicting Experiment

Directory of Open Access Journals (Sweden)

Jing Lu

2014-11-01

Full Text Available We propose a weather prediction model in this article based on neural network and fuzzy inference system (NFIS-WPM, and then apply it to predict daily fuzzy precipitation given meteorological premises for testing. The model consists of two parts: the first part is the “fuzzy rule-based neural network”, which simulates sequential relations among fuzzy sets using artificial neural network; and the second part is the “neural fuzzy inference system”, which is based on the first part, but could learn new fuzzy rules from the previous ones according to the algorithm we proposed. NFIS-WPM (High Pro and NFIS-WPM (Ave are improved versions of this model. It is well known that the need for accurate weather prediction is apparent when considering the benefits. However, the excessive pursuit of accuracy in weather prediction makes some of the “accurate” prediction results meaningless and the numerical prediction model is often complex and time-consuming. By adapting this novel model to a precipitation prediction problem, we make the predicted outcomes of precipitation more accurate and the prediction methods simpler than by using the complex numerical forecasting model that would occupy large computation resources, be time-consuming and which has a low predictive accuracy rate. Accordingly, we achieve more accurate predictive precipitation results than by using traditional artificial neural networks that have low predictive accuracy.

NOx PREDICTION FOR FBC BOILERS USING EMPIRICAL MODELS

Directory of Open Access Journals (Sweden)

Jiří Štefanica

2014-02-01

Full Text Available Reliable prediction of NOx emissions can provide useful information for boiler design and fuel selection. Recently used kinetic prediction models for FBC boilers are overly complex and require large computing capacity. Even so, there are many uncertainties in the case of FBC boilers. An empirical modeling approach for NOx prediction has been used exclusively for PCC boilers. No reference is available for modifying this method for FBC conditions. This paper presents possible advantages of empirical modeling based prediction of NOx emissions for FBC boilers, together with a discussion of its limitations. Empirical models are reviewed, and are applied to operation data from FBC boilers used for combusting Czech lignite coal or coal-biomass mixtures. Modifications to the model are proposed in accordance with theoretical knowledge and prediction accuracy.

The "AQUASCOPE" simplified model for predicting 89, 90Sr, 131l and 134, 137Cs in surface waters after a large-scale radioactive fallout

NARCIS (Netherlands)

Smith, J.T.; Belova, N.V.; Bulgakov, A.A.; Comans, R.N.J.; Konoplev, A.V.; Kudelsky, A.V.; Madruga, M.J.; Voitsekhovitch, O.V.; Zibolt, G.

2005-01-01

Simplified dynamic models have been developed for predicting the concentrations of radiocesium, radiostrontium, and 131I in surface waters and freshwater fish following a large-scale radioactive fallout. The models are intended to give averaged estimates for radionuclides in water bodies and in fish

Dynameomics: Data-driven methods and models for utilizing large-scale protein structure repositories for improving fragment-based loop prediction

Science.gov (United States)

Rysavy, Steven J; Beck, David AC; Daggett, Valerie

2014-01-01

Protein function is intimately linked to protein structure and dynamics yet experimentally determined structures frequently omit regions within a protein due to indeterminate data, which is often due protein dynamics. We propose that atomistic molecular dynamics simulations provide a diverse sampling of biologically relevant structures for these missing segments (and beyond) to improve structural modeling and structure prediction. Here we make use of the Dynameomics data warehouse, which contains simulations of representatives of essentially all known protein folds. We developed novel computational methods to efficiently identify, rank and retrieve small peptide structures, or fragments, from this database. We also created a novel data model to analyze and compare large repositories of structural data, such as contained within the Protein Data Bank and the Dynameomics data warehouse. Our evaluation compares these structural repositories for improving loop predictions and analyzes the utility of our methods and models. Using a standard set of loop structures, containing 510 loops, 30 for each loop length from 4 to 20 residues, we find that the inclusion of Dynameomics structures in fragment-based methods improves the quality of the loop predictions without being dependent on sequence homology. Depending on loop length, ∼25–75% of the best predictions came from the Dynameomics set, resulting in lower main chain root-mean-square deviations for all fragment lengths using the combined fragment library. We also provide specific cases where Dynameomics fragments provide better predictions for NMR loop structures than fragments from crystal structures. Online access to these fragment libraries is available at http://www.dynameomics.org/fragments. PMID:25142412

Dynameomics: data-driven methods and models for utilizing large-scale protein structure repositories for improving fragment-based loop prediction.

Science.gov (United States)

Rysavy, Steven J; Beck, David A C; Daggett, Valerie

2014-11-01

Protein function is intimately linked to protein structure and dynamics yet experimentally determined structures frequently omit regions within a protein due to indeterminate data, which is often due protein dynamics. We propose that atomistic molecular dynamics simulations provide a diverse sampling of biologically relevant structures for these missing segments (and beyond) to improve structural modeling and structure prediction. Here we make use of the Dynameomics data warehouse, which contains simulations of representatives of essentially all known protein folds. We developed novel computational methods to efficiently identify, rank and retrieve small peptide structures, or fragments, from this database. We also created a novel data model to analyze and compare large repositories of structural data, such as contained within the Protein Data Bank and the Dynameomics data warehouse. Our evaluation compares these structural repositories for improving loop predictions and analyzes the utility of our methods and models. Using a standard set of loop structures, containing 510 loops, 30 for each loop length from 4 to 20 residues, we find that the inclusion of Dynameomics structures in fragment-based methods improves the quality of the loop predictions without being dependent on sequence homology. Depending on loop length, ∼ 25-75% of the best predictions came from the Dynameomics set, resulting in lower main chain root-mean-square deviations for all fragment lengths using the combined fragment library. We also provide specific cases where Dynameomics fragments provide better predictions for NMR loop structures than fragments from crystal structures. Online access to these fragment libraries is available at http://www.dynameomics.org/fragments. © 2014 The Protein Society.

5D Modelling: An Efficient Approach for Creating Spatiotemporal Predictive 3D Maps of Large-Scale Cultural Resources

Science.gov (United States)

Doulamis, A.; Doulamis, N.; Ioannidis, C.; Chrysouli, C.; Grammalidis, N.; Dimitropoulos, K.; Potsiou, C.; Stathopoulou, E.-K.; Ioannides, M.

2015-08-01

Outdoor large-scale cultural sites are mostly sensitive to environmental, natural and human made factors, implying an imminent need for a spatio-temporal assessment to identify regions of potential cultural interest (material degradation, structuring, conservation). On the other hand, in Cultural Heritage research quite different actors are involved (archaeologists, curators, conservators, simple users) each of diverse needs. All these statements advocate that a 5D modelling (3D geometry plus time plus levels of details) is ideally required for preservation and assessment of outdoor large scale cultural sites, which is currently implemented as a simple aggregation of 3D digital models at different time and levels of details. The main bottleneck of such an approach is its complexity, making 5D modelling impossible to be validated in real life conditions. In this paper, a cost effective and affordable framework for 5D modelling is proposed based on a spatial-temporal dependent aggregation of 3D digital models, by incorporating a predictive assessment procedure to indicate which regions (surfaces) of an object should be reconstructed at higher levels of details at next time instances and which at lower ones. In this way, dynamic change history maps are created, indicating spatial probabilities of regions needed further 3D modelling at forthcoming instances. Using these maps, predictive assessment can be made, that is, to localize surfaces within the objects where a high accuracy reconstruction process needs to be activated at the forthcoming time instances. The proposed 5D Digital Cultural Heritage Model (5D-DCHM) is implemented using open interoperable standards based on the CityGML framework, which also allows the description of additional semantic metadata information. Visualization aspects are also supported to allow easy manipulation, interaction and representation of the 5D-DCHM geometry and the respective semantic information. The open source 3DCity

Neutron fraction and neutrino mean free path predictions in relativistic mean field models

International Nuclear Information System (INIS)

Hutauruk, P.T.P.; Williams, C.K.; Sulaksono, A.; Mart, T.

2004-01-01

The equation of state (EOS) of dense matter and neutrino mean free path (NMFP) in a neutron star have been studied by using relativistic mean field models motivated by effective field theory. It is found that the models predict too large proton fractions, although one of the models (G2) predicts an acceptable EOS. This is caused by the isovector terms. Except G2, the other two models predict anomalous NMFP's. In order to minimize the anomaly, besides an acceptable EOS, a large M* is favorable. A model with large M* retains the regularity in the NMFP even for a small neutron fraction

Prediction model of potential hepatocarcinogenicity of rat hepatocarcinogens using a large-scale toxicogenomics database

International Nuclear Information System (INIS)

Uehara, Takeki; Minowa, Yohsuke; Morikawa, Yuji; Kondo, Chiaki; Maruyama, Toshiyuki; Kato, Ikuo; Nakatsu, Noriyuki; Igarashi, Yoshinobu; Ono, Atsushi; Hayashi, Hitomi; Mitsumori, Kunitoshi; Yamada, Hiroshi; Ohno, Yasuo; Urushidani, Tetsuro

2011-01-01

The present study was performed to develop a robust gene-based prediction model for early assessment of potential hepatocarcinogenicity of chemicals in rats by using our toxicogenomics database, TG-GATEs (Genomics-Assisted Toxicity Evaluation System developed by the Toxicogenomics Project in Japan). The positive training set consisted of high- or middle-dose groups that received 6 different non-genotoxic hepatocarcinogens during a 28-day period. The negative training set consisted of high- or middle-dose groups of 54 non-carcinogens. Support vector machine combined with wrapper-type gene selection algorithms was used for modeling. Consequently, our best classifier yielded prediction accuracies for hepatocarcinogenicity of 99% sensitivity and 97% specificity in the training data set, and false positive prediction was almost completely eliminated. Pathway analysis of feature genes revealed that the mitogen-activated protein kinase p38- and phosphatidylinositol-3-kinase-centered interactome and the v-myc myelocytomatosis viral oncogene homolog-centered interactome were the 2 most significant networks. The usefulness and robustness of our predictor were further confirmed in an independent validation data set obtained from the public database. Interestingly, similar positive predictions were obtained in several genotoxic hepatocarcinogens as well as non-genotoxic hepatocarcinogens. These results indicate that the expression profiles of our newly selected candidate biomarker genes might be common characteristics in the early stage of carcinogenesis for both genotoxic and non-genotoxic carcinogens in the rat liver. Our toxicogenomic model might be useful for the prospective screening of hepatocarcinogenicity of compounds and prioritization of compounds for carcinogenicity testing. - Highlights: →We developed a toxicogenomic model to predict hepatocarcinogenicity of chemicals. →The optimized model consisting of 9 probes had 99% sensitivity and 97% specificity. �

Modeling Temporal Behavior in Large Networks: A Dynamic Mixed-Membership Model

Energy Technology Data Exchange (ETDEWEB)

Rossi, R; Gallagher, B; Neville, J; Henderson, K

2011-11-11

Given a large time-evolving network, how can we model and characterize the temporal behaviors of individual nodes (and network states)? How can we model the behavioral transition patterns of nodes? We propose a temporal behavior model that captures the 'roles' of nodes in the graph and how they evolve over time. The proposed dynamic behavioral mixed-membership model (DBMM) is scalable, fully automatic (no user-defined parameters), non-parametric/data-driven (no specific functional form or parameterization), interpretable (identifies explainable patterns), and flexible (applicable to dynamic and streaming networks). Moreover, the interpretable behavioral roles are generalizable, computationally efficient, and natively supports attributes. We applied our model for (a) identifying patterns and trends of nodes and network states based on the temporal behavior, (b) predicting future structural changes, and (c) detecting unusual temporal behavior transitions. We use eight large real-world datasets from different time-evolving settings (dynamic and streaming). In particular, we model the evolving mixed-memberships and the corresponding behavioral transitions of Twitter, Facebook, IP-Traces, Email (University), Internet AS, Enron, Reality, and IMDB. The experiments demonstrate the scalability, flexibility, and effectiveness of our model for identifying interesting patterns, detecting unusual structural transitions, and predicting the future structural changes of the network and individual nodes.

Predictive modeling of coupled multi-physics systems: I. Theory

International Nuclear Information System (INIS)

Cacuci, Dan Gabriel

2014-01-01

Highlights: • We developed “predictive modeling of coupled multi-physics systems (PMCMPS)”. • PMCMPS reduces predicted uncertainties in predicted model responses and parameters. • PMCMPS treats efficiently very large coupled systems. - Abstract: This work presents an innovative mathematical methodology for “predictive modeling of coupled multi-physics systems (PMCMPS).” This methodology takes into account fully the coupling terms between the systems but requires only the computational resources that would be needed to perform predictive modeling on each system separately. The PMCMPS methodology uses the maximum entropy principle to construct an optimal approximation of the unknown a priori distribution based on a priori known mean values and uncertainties characterizing the parameters and responses for both multi-physics models. This “maximum entropy”-approximate a priori distribution is combined, using Bayes’ theorem, with the “likelihood” provided by the multi-physics simulation models. Subsequently, the posterior distribution thus obtained is evaluated using the saddle-point method to obtain analytical expressions for the optimally predicted values for the multi-physics models parameters and responses along with corresponding reduced uncertainties. Noteworthy, the predictive modeling methodology for the coupled systems is constructed such that the systems can be considered sequentially rather than simultaneously, while preserving exactly the same results as if the systems were treated simultaneously. Consequently, very large coupled systems, which could perhaps exceed available computational resources if treated simultaneously, can be treated with the PMCMPS methodology presented in this work sequentially and without any loss of generality or information, requiring just the resources that would be needed if the systems were treated sequentially

Creep Rupture Life Prediction Based on Analysis of Large Creep Deformation

Directory of Open Access Journals (Sweden)

YE Wenming

2016-08-01

Full Text Available A creep rupture life prediction method for high temperature component was proposed. The method was based on a true stress-strain elastoplastic creep constitutive model and the large deformation finite element analysis method. This method firstly used the high-temperature tensile stress-strain curve expressed by true stress and strain and the creep curve to build materials' elastoplastic and creep constitutive model respectively, then used the large deformation finite element method to calculate the deformation response of high temperature component under a given load curve, finally the creep rupture life was determined according to the change trend of the responsive curve.The method was verified by durable test of TC11 titanium alloy notched specimens under 500 ℃, and was compared with the three creep rupture life prediction methods based on the small deformation analysis. Results show that the proposed method can accurately predict the high temperature creep response and long-term life of TC11 notched specimens, and the accuracy is better than that of the methods based on the average effective stress of notch ligament, the bone point stress and the fracture strain of the key point, which are all based on small deformation finite element analysis.

Toxicological relationships between proteins obtained from protein target predictions of large toxicity databases

International Nuclear Information System (INIS)

Nigsch, Florian; Mitchell, John B.O.

2008-01-01

The combination of models for protein target prediction with large databases containing toxicological information for individual molecules allows the derivation of 'toxiclogical' profiles, i.e., to what extent are molecules of known toxicity predicted to interact with a set of protein targets. To predict protein targets of drug-like and toxic molecules, we built a computational multiclass model using the Winnow algorithm based on a dataset of protein targets derived from the MDL Drug Data Report. A 15-fold Monte Carlo cross-validation using 50% of each class for training, and the remaining 50% for testing, provided an assessment of the accuracy of that model. We retained the 3 top-ranking predictions and found that in 82% of all cases the correct target was predicted within these three predictions. The first prediction was the correct one in almost 70% of cases. A model built on the whole protein target dataset was then used to predict the protein targets for 150 000 molecules from the MDL Toxicity Database. We analysed the frequency of the predictions across the panel of protein targets for experimentally determined toxicity classes of all molecules. This allowed us to identify clusters of proteins related by their toxicological profiles, as well as toxicities that are related. Literature-based evidence is provided for some specific clusters to show the relevance of the relationships identified

The Large-scale Coronal Structure of the 2017 August 21 Great American Eclipse: An Assessment of Solar Surface Flux Transport Model Enabled Predictions and Observations

Science.gov (United States)

Nandy, Dibyendu; Bhowmik, Prantika; Yeates, Anthony R.; Panda, Suman; Tarafder, Rajashik; Dash, Soumyaranjan

2018-01-01

On 2017 August 21, a total solar eclipse swept across the contiguous United States, providing excellent opportunities for diagnostics of the Sun’s corona. The Sun’s coronal structure is notoriously difficult to observe except during solar eclipses; thus, theoretical models must be relied upon for inferring the underlying magnetic structure of the Sun’s outer atmosphere. These models are necessary for understanding the role of magnetic fields in the heating of the corona to a million degrees and the generation of severe space weather. Here we present a methodology for predicting the structure of the coronal field based on model forward runs of a solar surface flux transport model, whose predicted surface field is utilized to extrapolate future coronal magnetic field structures. This prescription was applied to the 2017 August 21 solar eclipse. A post-eclipse analysis shows good agreement between model simulated and observed coronal structures and their locations on the limb. We demonstrate that slow changes in the Sun’s surface magnetic field distribution driven by long-term flux emergence and its evolution governs large-scale coronal structures with a (plausibly cycle-phase dependent) dynamical memory timescale on the order of a few solar rotations, opening up the possibility for large-scale, global corona predictions at least a month in advance.

Heave motion prediction of a large barge in random seas by using artificial neural network

Science.gov (United States)

Lee, Hsiu Eik; Liew, Mohd Shahir; Zawawi, Noor Amila Wan Abdullah; Toloue, Iraj

2017-11-01

This paper describes the development of a multi-layer feed forward artificial neural network (ANN) to predict rigid heave body motions of a large catenary moored barge subjected to multi-directional irregular waves. The barge is idealized as a rigid plate of finite draft with planar dimensions 160m (length) and 100m (width) which is held on station using a six point chain catenary mooring in 50m water depth. Hydroelastic effects are neglected from the physical model as the chief intent of this study is focused on large plate rigid body hydrodynamics modelling using ANN. Even with this assumption, the computational requirements for time domain coupled hydrodynamic simulations of a moored floating body is considerably costly, particularly if a large number of simulations are required such as in the case of response based design (RBD) methods. As an alternative to time consuming numerical hydrodynamics, a regression-type ANN model has been developed for efficient prediction of the barge's heave responses to random waves from various directions. It was determined that a network comprising of 3 input features, 2 hidden layers with 5 neurons each and 1 output was sufficient to produce acceptable predictions within 0.02 mean squared error. By benchmarking results from the ANN with those generated by a fully coupled dynamic model in OrcaFlex, it is demonstrated that the ANN is capable of predicting the barge's heave responses with acceptable accuracy.

From GenBank to GBIF: Phylogeny-Based Predictive Niche Modeling Tests Accuracy of Taxonomic Identifications in Large Occurrence Data Repositories.

Science.gov (United States)

Smith, B Eugene; Johnston, Mark K; Lücking, Robert

2016-01-01

Accuracy of taxonomic identifications is crucial to data quality in online repositories of species occurrence data, such as the Global Biodiversity Information Facility (GBIF), which have accumulated several hundred million records over the past 15 years. These data serve as basis for large scale analyses of macroecological and biogeographic patterns and to document environmental changes over time. However, taxonomic identifications are often unreliable, especially for non-vascular plants and fungi including lichens, which may lack critical revisions of voucher specimens. Due to the scale of the problem, restudy of millions of collections is unrealistic and other strategies are needed. Here we propose to use verified, georeferenced occurrence data of a given species to apply predictive niche modeling that can then be used to evaluate unverified occurrences of that species. Selecting the charismatic lichen fungus, Usnea longissima, as a case study, we used georeferenced occurrence records based on sequenced specimens to model its predicted niche. Our results suggest that the target species is largely restricted to a narrow range of boreal and temperate forest in the Northern Hemisphere and that occurrence records in GBIF from tropical regions and the Southern Hemisphere do not represent this taxon, a prediction tested by comparison with taxonomic revisions of Usnea for these regions. As a novel approach, we employed Principal Component Analysis on the environmental grid data used for predictive modeling to visualize potential ecogeographical barriers for the target species; we found that tropical regions conform a strong barrier, explaining why potential niches in the Southern Hemisphere were not colonized by Usnea longissima and instead by morphologically similar species. This approach is an example of how data from two of the most important biodiversity repositories, GenBank and GBIF, can be effectively combined to remotely address the problem of inaccuracy of

Predicting artificailly drained areas by means of selective model ensemble

DEFF Research Database (Denmark)

Møller, Anders Bjørn; Beucher, Amélie; Iversen, Bo Vangsø

. The approaches employed include decision trees, discriminant analysis, regression models, neural networks and support vector machines amongst others. Several models are trained with each method, using variously the original soil covariates and principal components of the covariates. With a large ensemble...... out since the mid-19th century, and it has been estimated that half of the cultivated area is artificially drained (Olesen, 2009). A number of machine learning approaches can be used to predict artificially drained areas in geographic space. However, instead of choosing the most accurate model....... The study aims firstly to train a large number of models to predict the extent of artificially drained areas using various machine learning approaches. Secondly, the study will develop a method for selecting the models, which give a good prediction of artificially drained areas, when used in conjunction...

Comprehensive Prediction of Large-height Swell-like Waves in East Coast of Korea

Science.gov (United States)

Kwon, S. J.; Lee, C.; Ahn, S. J.; Kim, H. K.

2014-12-01

There have been growing interests in the large-height swell-like wave (LSW) in the east coast of Korea because such big waves have caused human victims as well as damages to facilities such as breakwaters in the coast. The LSW was found to be generated due to an atmospherically great valley in the north area of the East Sea and then propagate long distance to the east coast of Korea in prominently southwest direction (Oh et al., 2010).In this study, we will perform two methods, real-time data based and numerical-model based predictions in order to predict the LSW in the east coast of Korea. First, the real-time data based prediction method uses information which is collected by the directional wave gauge installed near Sokcho. Using the wave model SWAN (Booij et al., 1999) and the wave ray method (Munk and Arthur, 1952), we will estimate wave data in open sea from the real-time data and predict the travel time of LSW from the measurement site (near Sokcho) to several target points in the east coast of Korea. Second, the numerical-model based method uses three different numerical models; WW3 in deep water, SWAN in shallow water, and CADMAS-SURF for wave run-up (CDIT). The surface winds from the 72 hours prediction system of NCEP (National Centers for Environmental Prediction) GFS (Global Forecast System) will be inputted in finer grids after interpolating these in certain domains of WW3 and SWAN models. The significant wave heights and peak wave directions predicted by the two methods will be compared to the measured data of LSW at several target points near the coasts. Further, the prediction method will be improved using more measurement sites which will be installed in the future. ReferencesBooij, N., Ris, R.C., and Holthuijsen, L.H. (1999). A third-generation wave model for coastal regions 1. Model description and validation. J. of Geophysical Research, 103(C4), 7649-7666.Munk, W.H. and Arthur, R.S. (1952). Gravity Waves. 13. Wave Intensity along a Refracted Ray

Seasonal predictability of Kiremt rainfall in coupled general circulation models

Science.gov (United States)

Gleixner, Stephanie; Keenlyside, Noel S.; Demissie, Teferi D.; Counillon, François; Wang, Yiguo; Viste, Ellen

2017-11-01

The Ethiopian economy and population is strongly dependent on rainfall. Operational seasonal predictions for the main rainy season (Kiremt, June-September) are based on statistical approaches with Pacific sea surface temperatures (SST) as the main predictor. Here we analyse dynamical predictions from 11 coupled general circulation models for the Kiremt seasons from 1985-2005 with the forecasts starting from the beginning of May. We find skillful predictions from three of the 11 models, but no model beats a simple linear prediction model based on the predicted Niño3.4 indices. The skill of the individual models for dynamically predicting Kiremt rainfall depends on the strength of the teleconnection between Kiremt rainfall and concurrent Pacific SST in the models. Models that do not simulate this teleconnection fail to capture the observed relationship between Kiremt rainfall and the large-scale Walker circulation.

Improving Predictive Modeling in Pediatric Drug Development: Pharmacokinetics, Pharmacodynamics, and Mechanistic Modeling

Energy Technology Data Exchange (ETDEWEB)

Slikker, William; Young, John F.; Corley, Rick A.; Dorman, David C.; Conolly, Rory B.; Knudsen, Thomas; Erstad, Brian L.; Luecke, Richard H.; Faustman, Elaine M.; Timchalk, Chuck; Mattison, Donald R.

2005-07-26

A workshop was conducted on November 18?19, 2004, to address the issue of improving predictive models for drug delivery to developing humans. Although considerable progress has been made for adult humans, large gaps remain for predicting pharmacokinetic/pharmacodynamic (PK/PD) outcome in children because most adult models have not been tested during development. The goals of the meeting included a description of when, during development, infants/children become adultlike in handling drugs. The issue of incorporating the most recent advances into the predictive models was also addressed: both the use of imaging approaches and genomic information were considered. Disease state, as exemplified by obesity, was addressed as a modifier of drug pharmacokinetics and pharmacodynamics during development. Issues addressed in this workshop should be considered in the development of new predictive and mechanistic models of drug kinetics and dynamics in the developing human.

Distributed Model Predictive Control for Smart Energy Systems

DEFF Research Database (Denmark)

Halvgaard, Rasmus Fogtmann; Vandenberghe, Lieven; Poulsen, Niels Kjølstad

2016-01-01

Integration of a large number of flexible consumers in a smart grid requires a scalable power balancing strategy. We formulate the control problem as an optimization problem to be solved repeatedly by the aggregator in a model predictive control framework. To solve the large-scale control problem...

Scalable Joint Models for Reliable Uncertainty-Aware Event Prediction.

Science.gov (United States)

Soleimani, Hossein; Hensman, James; Saria, Suchi

2017-08-21

Missing data and noisy observations pose significant challenges for reliably predicting events from irregularly sampled multivariate time series (longitudinal) data. Imputation methods, which are typically used for completing the data prior to event prediction, lack a principled mechanism to account for the uncertainty due to missingness. Alternatively, state-of-the-art joint modeling techniques can be used for jointly modeling the longitudinal and event data and compute event probabilities conditioned on the longitudinal observations. These approaches, however, make strong parametric assumptions and do not easily scale to multivariate signals with many observations. Our proposed approach consists of several key innovations. First, we develop a flexible and scalable joint model based upon sparse multiple-output Gaussian processes. Unlike state-of-the-art joint models, the proposed model can explain highly challenging structure including non-Gaussian noise while scaling to large data. Second, we derive an optimal policy for predicting events using the distribution of the event occurrence estimated by the joint model. The derived policy trades-off the cost of a delayed detection versus incorrect assessments and abstains from making decisions when the estimated event probability does not satisfy the derived confidence criteria. Experiments on a large dataset show that the proposed framework significantly outperforms state-of-the-art techniques in event prediction.

Prediction of broadband trailing edge noise from a NACA0012 airfoil using wall-modeled large-eddy simulation

Science.gov (United States)

Mehrabadi, Mohammad; Bodony, Daniel

2017-11-01

In modern high-bypass ratio turbofan engines, the reduction of jet exhaust noise through engine design has increased the acoustic importance of the main fan to the point where it can be the primary source of noise in the fight direction of an airplane. While fan noise has been reduced by improved fan designs, its broadband component, originating from the interaction of turbulent flow with a solid surface, still remains an issue. Broadband fan noise is generated by several mechanisms, usually involving a turbulent boundary layer interacting with a solid surface. To prepare for a wall modeled large eddy simulation (WMLES) of the NASA/GE source diagnostic test fan, we study the broadband noise due to the turbulent flow on a NACA0012 airfoil at zero degree angle-of-attack, a chord-based Reynolds number of 408,000, and a Mach number of 0.115 using WMLES. We investigate the prediction of transition-to-turbulence and sound generation from the WMLES and examine its predictability compared with available experimental and DNS datasets for the same flow conditions. Verification of WMLES for such a canonical problem is crucial since it provides useful insight about the WMLES approach before using it for broadband fan noise prediction. AeroAcoustics Research Consortium.

Comparison of two ordinal prediction models

DEFF Research Database (Denmark)

Kattan, Michael W; Gerds, Thomas A

2015-01-01

system (i.e. old or new), such as the level of evidence for one or more factors included in the system or the general opinions of expert clinicians. However, given the major objective of estimating prognosis on an ordinal scale, we argue that the rival staging system candidates should be compared...... on their ability to predict outcome. We sought to outline an algorithm that would compare two rival ordinal systems on their predictive ability. RESULTS: We devised an algorithm based largely on the concordance index, which is appropriate for comparing two models in their ability to rank observations. We...... demonstrate our algorithm with a prostate cancer staging system example. CONCLUSION: We have provided an algorithm for selecting the preferred staging system based on prognostic accuracy. It appears to be useful for the purpose of selecting between two ordinal prediction models....

Modelling and measurements of wakes in large wind farms

International Nuclear Information System (INIS)

Barthelmie, R J; Rathmann, O; Frandsen, S T; Hansen, K S; Politis, E; Prospathopoulos, J; Rados, K; Cabezon, D; Schlez, W; Phillips, J; Neubert, A; Schepers, J G; Pijl, S P van der

2007-01-01

The paper presents research conducted in the Flow workpackage of the EU funded UPWIND project which focuses on improving models of flow within and downwind of large wind farms in complex terrain and offshore. The main activity is modelling the behaviour of wind turbine wakes in order to improve power output predictions

Stabilizing model predictive control : on the enlargement of the terminal set

NARCIS (Netherlands)

Brunner, F.D.; Lazar, M.; Allgöwer, F.

2015-01-01

It is well known that a large terminal set leads to a large region where the model predictive control problem is feasible without the need for a long prediction horizon. This paper proposes a new method for the enlargement of the terminal set. Different from existing approaches, the method uses the

Bootstrap prediction and Bayesian prediction under misspecified models

OpenAIRE

Fushiki, Tadayoshi

2005-01-01

We consider a statistical prediction problem under misspecified models. In a sense, Bayesian prediction is an optimal prediction method when an assumed model is true. Bootstrap prediction is obtained by applying Breiman's `bagging' method to a plug-in prediction. Bootstrap prediction can be considered to be an approximation to the Bayesian prediction under the assumption that the model is true. However, in applications, there are frequently deviations from the assumed model. In this paper, bo...

Nudging and predictability in regional climate modelling: investigation in a nested quasi-geostrophic model

Science.gov (United States)

Omrani, Hiba; Drobinski, Philippe; Dubos, Thomas

2010-05-01

In this work, we consider the effect of indiscriminate and spectral nudging on the large and small scales of an idealized model simulation. The model is a two layer quasi-geostrophic model on the beta-plane driven at its boundaries by the « global » version with periodic boundary condition. This setup mimics the configuration used for regional climate modelling. The effect of large-scale nudging is studied by using the "perfect model" approach. Two sets of experiments are performed: (1) the effect of nudging is investigated with a « global » high resolution two layer quasi-geostrophic model driven by a low resolution two layer quasi-geostrophic model. (2) similar simulations are conducted with the two layer quasi-geostrophic Limited Area Model (LAM) where the size of the LAM domain comes into play in addition to the first set of simulations. The study shows that the indiscriminate nudging time that minimizes the error at both the large and small scales is reached for a nudging time close to the predictability time, for spectral nudging, the optimum nudging time should tend to zero since the best large scale dynamics is supposed to be given by the driving large-scale fields are generally given at much lower frequency than the model time step(e,g, 6-hourly analysis) with a basic interpolation between the fields, the optimum nudging time differs from zero, however remaining smaller than the predictability time.

Falling in the elderly: Do statistical models matter for performance criteria of fall prediction? Results from two large population-based studies.

Science.gov (United States)

Kabeshova, Anastasiia; Launay, Cyrille P; Gromov, Vasilii A; Fantino, Bruno; Levinoff, Elise J; Allali, Gilles; Beauchet, Olivier

2016-01-01

To compare performance criteria (i.e., sensitivity, specificity, positive predictive value, negative predictive value, area under receiver operating characteristic curve and accuracy) of linear and non-linear statistical models for fall risk in older community-dwellers. Participants were recruited in two large population-based studies, "Prévention des Chutes, Réseau 4" (PCR4, n=1760, cross-sectional design, retrospective collection of falls) and "Prévention des Chutes Personnes Agées" (PCPA, n=1765, cohort design, prospective collection of falls). Six linear statistical models (i.e., logistic regression, discriminant analysis, Bayes network algorithm, decision tree, random forest, boosted trees), three non-linear statistical models corresponding to artificial neural networks (multilayer perceptron, genetic algorithm and neuroevolution of augmenting topologies [NEAT]) and the adaptive neuro fuzzy interference system (ANFIS) were used. Falls ≥1 characterizing fallers and falls ≥2 characterizing recurrent fallers were used as outcomes. Data of studies were analyzed separately and together. NEAT and ANFIS had better performance criteria compared to other models. The highest performance criteria were reported with NEAT when using PCR4 database and falls ≥1, and with both NEAT and ANFIS when pooling data together and using falls ≥2. However, sensitivity and specificity were unbalanced. Sensitivity was higher than specificity when identifying fallers, whereas the converse was found when predicting recurrent fallers. Our results showed that NEAT and ANFIS were non-linear statistical models with the best performance criteria for the prediction of falls but their sensitivity and specificity were unbalanced, underscoring that models should be used respectively for the screening of fallers and the diagnosis of recurrent fallers. Copyright © 2015 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.

Prediction skill of rainstorm events over India in the TIGGE weather prediction models

Science.gov (United States)

Karuna Sagar, S.; Rajeevan, M.; Vijaya Bhaskara Rao, S.; Mitra, A. K.

2017-12-01

Extreme rainfall events pose a serious threat of leading to severe floods in many countries worldwide. Therefore, advance prediction of its occurrence and spatial distribution is very essential. In this paper, an analysis has been made to assess the skill of numerical weather prediction models in predicting rainstorms over India. Using gridded daily rainfall data set and objective criteria, 15 rainstorms were identified during the monsoon season (June to September). The analysis was made using three TIGGE (THe Observing System Research and Predictability Experiment (THORPEX) Interactive Grand Global Ensemble) models. The models considered are the European Centre for Medium-Range Weather Forecasts (ECMWF), National Centre for Environmental Prediction (NCEP) and the UK Met Office (UKMO). Verification of the TIGGE models for 43 observed rainstorm days from 15 rainstorm events has been made for the period 2007-2015. The comparison reveals that rainstorm events are predictable up to 5 days in advance, however with a bias in spatial distribution and intensity. The statistical parameters like mean error (ME) or Bias, root mean square error (RMSE) and correlation coefficient (CC) have been computed over the rainstorm region using the multi-model ensemble (MME) mean. The study reveals that the spread is large in ECMWF and UKMO followed by the NCEP model. Though the ensemble spread is quite small in NCEP, the ensemble member averages are not well predicted. The rank histograms suggest that the forecasts are under prediction. The modified Contiguous Rain Area (CRA) technique was used to verify the spatial as well as the quantitative skill of the TIGGE models. Overall, the contribution from the displacement and pattern errors to the total RMSE is found to be more in magnitude. The volume error increases from 24 hr forecast to 48 hr forecast in all the three models.

Use of models in large-area forest surveys: comparing model-assisted, model-based and hybrid estimation

Science.gov (United States)

Goran Stahl; Svetlana Saarela; Sebastian Schnell; Soren Holm; Johannes Breidenbach; Sean P. Healey; Paul L. Patterson; Steen Magnussen; Erik Naesset; Ronald E. McRoberts; Timothy G. Gregoire

2016-01-01

This paper focuses on the use of models for increasing the precision of estimators in large-area forest surveys. It is motivated by the increasing availability of remotely sensed data, which facilitates the development of models predicting the variables of interest in forest surveys. We present, review and compare three different estimation frameworks where...

Rare Plants of Southeastern Hardwood Forests and the Role of Predictive Modeling

International Nuclear Information System (INIS)

Imm, D.W.; Shealy, H.E. Jr.; McLeod, K.W.; Collins, B.

2001-01-01

Habitat prediction models for rare plants can be useful when large areas must be surveyed or populations must be established. Investigators developed a habitat prediction model for four species of Southeastern hardwood forests. These four examples suggest that models based on resource and vegetation characteristics can accurately predict habitat, but only when plants are strongly associated with these variables and the scale of modeling coincides with habitat size

Development and evaluation of a prediction model for underestimated invasive breast cancer in women with ductal carcinoma in situ at stereotactic large core needle biopsy.

Directory of Open Access Journals (Sweden)

Suzanne C E Diepstraten

Full Text Available BACKGROUND: We aimed to develop a multivariable model for prediction of underestimated invasiveness in women with ductal carcinoma in situ at stereotactic large core needle biopsy, that can be used to select patients for sentinel node biopsy at primary surgery. METHODS: From the literature, we selected potential preoperative predictors of underestimated invasive breast cancer. Data of patients with nonpalpable breast lesions who were diagnosed with ductal carcinoma in situ at stereotactic large core needle biopsy, drawn from the prospective COBRA (Core Biopsy after RAdiological localization and COBRA2000 cohort studies, were used to fit the multivariable model and assess its overall performance, discrimination, and calibration. RESULTS: 348 women with large core needle biopsy-proven ductal carcinoma in situ were available for analysis. In 100 (28.7% patients invasive carcinoma was found at subsequent surgery. Nine predictors were included in the model. In the multivariable analysis, the predictors with the strongest association were lesion size (OR 1.12 per cm, 95% CI 0.98-1.28, number of cores retrieved at biopsy (OR per core 0.87, 95% CI 0.75-1.01, presence of lobular cancerization (OR 5.29, 95% CI 1.25-26.77, and microinvasion (OR 3.75, 95% CI 1.42-9.87. The overall performance of the multivariable model was poor with an explained variation of 9% (Nagelkerke's R(2, mediocre discrimination with area under the receiver operating characteristic curve of 0.66 (95% confidence interval 0.58-0.73, and fairly good calibration. CONCLUSION: The evaluation of our multivariable prediction model in a large, clinically representative study population proves that routine clinical and pathological variables are not suitable to select patients with large core needle biopsy-proven ductal carcinoma in situ for sentinel node biopsy during primary surgery.

Predictive Modelling and Time: An Experiment in Temporal Archaeological Predictive Models

OpenAIRE

David Ebert

2006-01-01

One of the most common criticisms of archaeological predictive modelling is that it fails to account for temporal or functional differences in sites. However, a practical solution to temporal or functional predictive modelling has proven to be elusive. This article discusses temporal predictive modelling, focusing on the difficulties of employing temporal variables, then introduces and tests a simple methodology for the implementation of temporal modelling. The temporal models thus created ar...

Consensus models to predict endocrine disruption for all ...

Science.gov (United States)

Humans are potentially exposed to tens of thousands of man-made chemicals in the environment. It is well known that some environmental chemicals mimic natural hormones and thus have the potential to be endocrine disruptors. Most of these environmental chemicals have never been tested for their ability to disrupt the endocrine system, in particular, their ability to interact with the estrogen receptor. EPA needs tools to prioritize thousands of chemicals, for instance in the Endocrine Disruptor Screening Program (EDSP). Collaborative Estrogen Receptor Activity Prediction Project (CERAPP) was intended to be a demonstration of the use of predictive computational models on HTS data including ToxCast and Tox21 assays to prioritize a large chemical universe of 32464 unique structures for one specific molecular target – the estrogen receptor. CERAPP combined multiple computational models for prediction of estrogen receptor activity, and used the predicted results to build a unique consensus model. Models were developed in collaboration between 17 groups in the U.S. and Europe and applied to predict the common set of chemicals. Structure-based techniques such as docking and several QSAR modeling approaches were employed, mostly using a common training set of 1677 compounds provided by U.S. EPA, to build a total of 42 classification models and 8 regression models for binding, agonist and antagonist activity. All predictions were evaluated on ToxCast data and on an exte

SAS-macros for estimation and prediction in an model of the electricity consumption

DEFF Research Database (Denmark)

1998-01-01

SAS-macros for estimation and prediction in an model of the electricity consumption'' is a large collection of SAS-macros for handling a model of the electricity consumption in the Eastern Denmark. The macros are installed at Elkraft, Ballerup.......SAS-macros for estimation and prediction in an model of the electricity consumption'' is a large collection of SAS-macros for handling a model of the electricity consumption in the Eastern Denmark. The macros are installed at Elkraft, Ballerup....

Forced synchronization of large-scale circulation to increase predictability of surface states

Science.gov (United States)

Shen, Mao-Lin; Keenlyside, Noel; Selten, Frank; Wiegerinck, Wim; Duane, Gregory

2016-04-01

Numerical models are key tools in the projection of the future climate change. The lack of perfect initial condition and perfect knowledge of the laws of physics, as well as inherent chaotic behavior limit predictions. Conceptually, the atmospheric variables can be decomposed into a predictable component (signal) and an unpredictable component (noise). In ensemble prediction the anomaly of ensemble mean is regarded as the signal and the ensemble spread the noise. Naturally the prediction skill will be higher if the signal-to-noise ratio (SNR) is larger in the initial conditions. We run two ensemble experiments in order to explore a way to reduce the SNR of surface winds and temperature. One ensemble experiment is AGCM with prescribing sea surface temperature (SST); the other is AGCM with both prescribing SST and nudging the high-level temperature and winds to ERA-Interim. Each ensemble has 30 members. Larger SNR is expected and found over the tropical ocean in the first experiment because the tropical circulation is associated with the convection and the associated surface wind convergence as these are to a large extent driven by the SST. However, small SNR is found over high latitude ocean and land surface due to the chaotic and non-synchronized atmosphere states. In the second experiment the higher level temperature and winds are forced to be synchronized (nudged to reanalysis) and hence a larger SNR of surface winds and temperature is expected. Furthermore, different nudging coefficients are also tested in order to understand the limitation of both synchronization of large-scale circulation and the surface states. These experiments will be useful for the developing strategies to synchronize the 3-D states of atmospheric models that can be later used to build a super model.

Large-scale evaluation of dynamically important residues in proteins predicted by the perturbation analysis of a coarse-grained elastic model

Directory of Open Access Journals (Sweden)

Tekpinar Mustafa

2009-07-01

Full Text Available Abstract Backgrounds It is increasingly recognized that protein functions often require intricate conformational dynamics, which involves a network of key amino acid residues that couple spatially separated functional sites. Tremendous efforts have been made to identify these key residues by experimental and computational means. Results We have performed a large-scale evaluation of the predictions of dynamically important residues by a variety of computational protocols including three based on the perturbation and correlation analysis of a coarse-grained elastic model. This study is performed for two lists of test cases with >500 pairs of protein structures. The dynamically important residues predicted by the perturbation and correlation analysis are found to be strongly or moderately conserved in >67% of test cases. They form a sparse network of residues which are clustered both in 3D space and along protein sequence. Their overall conservation is attributed to their dynamic role rather than ligand binding or high network connectivity. Conclusion By modeling how the protein structural fluctuations respond to residue-position-specific perturbations, our highly efficient perturbation and correlation analysis can be used to dissect the functional conformational changes in various proteins with a residue level of detail. The predictions of dynamically important residues serve as promising targets for mutational and functional studies.

Zebrafish whole-adult-organism chemogenomics for large-scale predictive and discovery chemical biology.

Directory of Open Access Journals (Sweden)

Siew Hong Lam

2008-07-01

Full Text Available The ability to perform large-scale, expression-based chemogenomics on whole adult organisms, as in invertebrate models (worm and fly, is highly desirable for a vertebrate model but its feasibility and potential has not been demonstrated. We performed expression-based chemogenomics on the whole adult organism of a vertebrate model, the zebrafish, and demonstrated its potential for large-scale predictive and discovery chemical biology. Focusing on two classes of compounds with wide implications to human health, polycyclic (halogenated aromatic hydrocarbons [P(HAHs] and estrogenic compounds (ECs, we generated robust prediction models that can discriminate compounds of the same class from those of different classes in two large independent experiments. The robust expression signatures led to the identification of biomarkers for potent aryl hydrocarbon receptor (AHR and estrogen receptor (ER agonists, respectively, and were validated in multiple targeted tissues. Knowledge-based data mining of human homologs of zebrafish genes revealed highly conserved chemical-induced biological responses/effects, health risks, and novel biological insights associated with AHR and ER that could be inferred to humans. Thus, our study presents an effective, high-throughput strategy of capturing molecular snapshots of chemical-induced biological states of a whole adult vertebrate that provides information on biomarkers of effects, deregulated signaling pathways, and possible affected biological functions, perturbed physiological systems, and increased health risks. These findings place zebrafish in a strategic position to bridge the wide gap between cell-based and rodent models in chemogenomics research and applications, especially in preclinical drug discovery and toxicology.

Modeling of Complex Life Cycle Prediction Based on Cell Division

Directory of Open Access Journals (Sweden)

Fucheng Zhang

2017-01-01

Full Text Available Effective fault diagnosis and reasonable life expectancy are of great significance and practical engineering value for the safety, reliability, and maintenance cost of equipment and working environment. At present, the life prediction methods of the equipment are equipment life prediction based on condition monitoring, combined forecasting model, and driven data. Most of them need to be based on a large amount of data to achieve the problem. For this issue, we propose learning from the mechanism of cell division in the organism. We have established a moderate complexity of life prediction model across studying the complex multifactor correlation life model. In this paper, we model the life prediction of cell division. Experiments show that our model can effectively simulate the state of cell division. Through the model of reference, we will use it for the equipment of the complex life prediction.

Dark Radiation predictions from general Large Volume Scenarios

Science.gov (United States)

Hebecker, Arthur; Mangat, Patrick; Rompineve, Fabrizio; Witkowski, Lukas T.

2014-09-01

Recent observations constrain the amount of Dark Radiation (Δ N eff ) and may even hint towards a non-zero value of Δ N eff . It is by now well-known that this puts stringent constraints on the sequestered Large Volume Scenario (LVS), i.e. on LVS realisations with the Standard Model at a singularity. We go beyond this setting by considering LVS models where SM fields are realised on 7-branes in the geometric regime. As we argue, this naturally goes together with high-scale supersymmetry. The abundance of Dark Radiation is determined by the competition between the decay of the lightest modulus to axions, to the SM Higgs and to gauge fields, and leads to strict constraints on these models. Nevertheless, these constructions can in principle meet current DR bounds due to decays into gauge bosons alone. Further, a rather robust prediction for a substantial amount of Dark Radiation can be made. This applies both to cases where the SM 4-cycles are stabilised by D-terms and are small `by accident', i.e. tuning, as well as to fibred models with the small cycles stabilised by loops. In these constructions the DR axion and the QCD axion are the same field and we require a tuning of the initial misalignment to avoid Dark Matter overproduction. Furthermore, we analyse a closely related setting where the SM lives at a singularity but couples to the volume modulus through flavour branes. We conclude that some of the most natural LVS settings with natural values of model parameters lead to Dark Radiation predictions just below the present observational limits. Barring a discovery, rather modest improvements of present Dark Radiation bounds can rule out many of these most simple and generic variants of the LVS.

A multivariate model for predicting segmental body composition.

Science.gov (United States)

Tian, Simiao; Mioche, Laurence; Denis, Jean-Baptiste; Morio, Béatrice

2013-12-01

The aims of the present study were to propose a multivariate model for predicting simultaneously body, trunk and appendicular fat and lean masses from easily measured variables and to compare its predictive capacity with that of the available univariate models that predict body fat percentage (BF%). The dual-energy X-ray absorptiometry (DXA) dataset (52% men and 48% women) with White, Black and Hispanic ethnicities (1999-2004, National Health and Nutrition Examination Survey) was randomly divided into three sub-datasets: a training dataset (TRD), a test dataset (TED); a validation dataset (VAD), comprising 3835, 1917 and 1917 subjects. For each sex, several multivariate prediction models were fitted from the TRD using age, weight, height and possibly waist circumference. The most accurate model was selected from the TED and then applied to the VAD and a French DXA dataset (French DB) (526 men and 529 women) to assess the prediction accuracy in comparison with that of five published univariate models, for which adjusted formulas were re-estimated using the TRD. Waist circumference was found to improve the prediction accuracy, especially in men. For BF%, the standard error of prediction (SEP) values were 3.26 (3.75) % for men and 3.47 (3.95)% for women in the VAD (French DB), as good as those of the adjusted univariate models. Moreover, the SEP values for the prediction of body and appendicular lean masses ranged from 1.39 to 2.75 kg for both the sexes. The prediction accuracy was best for age < 65 years, BMI < 30 kg/m2 and the Hispanic ethnicity. The application of our multivariate model to large populations could be useful to address various public health issues.

Computational Modeling of Large Wildfires: A Roadmap

KAUST Repository

Coen, Janice L.

2010-08-01

Wildland fire behavior, particularly that of large, uncontrolled wildfires, has not been well understood or predicted. Our methodology to simulate this phenomenon uses high-resolution dynamic models made of numerical weather prediction (NWP) models coupled to fire behavior models to simulate fire behavior. NWP models are capable of modeling very high resolution (< 100 m) atmospheric flows. The wildland fire component is based upon semi-empirical formulas for fireline rate of spread, post-frontal heat release, and a canopy fire. The fire behavior is coupled to the atmospheric model such that low level winds drive the spread of the surface fire, which in turn releases sensible heat, latent heat, and smoke fluxes into the lower atmosphere, feeding back to affect the winds directing the fire. These coupled dynamic models capture the rapid spread downwind, flank runs up canyons, bifurcations of the fire into two heads, and rough agreement in area, shape, and direction of spread at periods for which fire location data is available. Yet, intriguing computational science questions arise in applying such models in a predictive manner, including physical processes that span a vast range of scales, processes such as spotting that cannot be modeled deterministically, estimating the consequences of uncertainty, the efforts to steer simulations with field data ("data assimilation"), lingering issues with short term forecasting of weather that may show skill only on the order of a few hours, and the difficulty of gathering pertinent data for verification and initialization in a dangerous environment. © 2010 IEEE.

A consensus approach for estimating the predictive accuracy of dynamic models in biology.

Science.gov (United States)

Villaverde, Alejandro F; Bongard, Sophia; Mauch, Klaus; Müller, Dirk; Balsa-Canto, Eva; Schmid, Joachim; Banga, Julio R

2015-04-01

Mathematical models that predict the complex dynamic behaviour of cellular networks are fundamental in systems biology, and provide an important basis for biomedical and biotechnological applications. However, obtaining reliable predictions from large-scale dynamic models is commonly a challenging task due to lack of identifiability. The present work addresses this challenge by presenting a methodology for obtaining high-confidence predictions from dynamic models using time-series data. First, to preserve the complex behaviour of the network while reducing the number of estimated parameters, model parameters are combined in sets of meta-parameters, which are obtained from correlations between biochemical reaction rates and between concentrations of the chemical species. Next, an ensemble of models with different parameterizations is constructed and calibrated. Finally, the ensemble is used for assessing the reliability of model predictions by defining a measure of convergence of model outputs (consensus) that is used as an indicator of confidence. We report results of computational tests carried out on a metabolic model of Chinese Hamster Ovary (CHO) cells, which are used for recombinant protein production. Using noisy simulated data, we find that the aggregated ensemble predictions are on average more accurate than the predictions of individual ensemble models. Furthermore, ensemble predictions with high consensus are statistically more accurate than ensemble predictions with large variance. The procedure provides quantitative estimates of the confidence in model predictions and enables the analysis of sufficiently complex networks as required for practical applications. Copyright © 2015 Elsevier Ireland Ltd. All rights reserved.

Attribution of Large-Scale Climate Patterns to Seasonal Peak-Flow and Prospects for Prediction Globally

Science.gov (United States)

Lee, Donghoon; Ward, Philip; Block, Paul

2018-02-01

Flood-related fatalities and impacts on society surpass those from all other natural disasters globally. While the inclusion of large-scale climate drivers in streamflow (or high-flow) prediction has been widely studied, an explicit link to global-scale long-lead prediction is lacking, which can lead to an improved understanding of potential flood propensity. Here we attribute seasonal peak-flow to large-scale climate patterns, including the El Niño Southern Oscillation (ENSO), Pacific Decadal Oscillation (PDO), North Atlantic Oscillation (NAO), and Atlantic Multidecadal Oscillation (AMO), using streamflow station observations and simulations from PCR-GLOBWB, a global-scale hydrologic model. Statistically significantly correlated climate patterns and streamflow autocorrelation are subsequently applied as predictors to build a global-scale season-ahead prediction model, with prediction performance evaluated by the mean squared error skill score (MSESS) and the categorical Gerrity skill score (GSS). Globally, fair-to-good prediction skill (20% ≤ MSESS and 0.2 ≤ GSS) is evident for a number of locations (28% of stations and 29% of land area), most notably in data-poor regions (e.g., West and Central Africa). The persistence of such relevant climate patterns can improve understanding of the propensity for floods at the seasonal scale. The prediction approach developed here lays the groundwork for further improving local-scale seasonal peak-flow prediction by identifying relevant global-scale climate patterns. This is especially attractive for regions with limited observations and or little capacity to develop flood early warning systems.

Two-group modeling of interfacial area transport in large diameter channels

Energy Technology Data Exchange (ETDEWEB)

Schlegel, J.P., E-mail: schlegelj@mst.edu [Department of Mining and Nuclear Engineering, Missouri University of Science and Technology, 301 W 14th St., Rolla, MO 65409 (United States); Hibiki, T.; Ishii, M. [School of Nuclear Engineering, Purdue University, 400 Central Dr., West Lafayette, IN 47907 (United States)

2015-11-15

Highlights: • Implemented updated constitutive models and benchmarking method for IATE in large pipes. • New model and method with new data improved the overall IATE prediction for large pipes. • Not all conditions well predicted shows that further development is still required. - Abstract: A comparison of the existing two-group interfacial area transport equation source and sink terms for large diameter channels with recently collected interfacial area concentration measurements (Schlegel et al., 2012, 2014. Int. J. Heat Fluid Flow 47, 42) has indicated that the model does not perform well in predicting interfacial area transport outside of the range of flow conditions used in the original benchmarking effort. In order to reduce the error in the prediction of interfacial area concentration by the interfacial area transport equation, several constitutive relations have been updated including the turbulence model and relative velocity correlation. The transport equation utilizing these updated models has been modified by updating the inter-group transfer and Group 2 coalescence and disintegration kernels using an expanded range of experimental conditions extending to pipe sizes of 0.304 m [12 in.], gas velocities of up to nearly 11 m/s [36.1 ft/s] and liquid velocities of up to 2 m/s [6.56 ft/s], as well as conditions with both bubbly flow and cap-bubbly flow injection (Schlegel et al., 2012, 2014). The modifications to the transport equation have resulted in a decrease in the RMS error for void fraction and interfacial area concentration from 17.32% to 12.3% and 21.26% to 19.6%. The combined RMS error, for both void fraction and interfacial area concentration, is below 15% for most of the experiments used in the comparison, a distinct improvement over the previous version of the model.

Risk Prediction Model for Severe Postoperative Complication in Bariatric Surgery.

Science.gov (United States)

Stenberg, Erik; Cao, Yang; Szabo, Eva; Näslund, Erik; Näslund, Ingmar; Ottosson, Johan

2018-01-12

Factors associated with risk for adverse outcome are important considerations in the preoperative assessment of patients for bariatric surgery. As yet, prediction models based on preoperative risk factors have not been able to predict adverse outcome sufficiently. This study aimed to identify preoperative risk factors and to construct a risk prediction model based on these. Patients who underwent a bariatric surgical procedure in Sweden between 2010 and 2014 were identified from the Scandinavian Obesity Surgery Registry (SOReg). Associations between preoperative potential risk factors and severe postoperative complications were analysed using a logistic regression model. A multivariate model for risk prediction was created and validated in the SOReg for patients who underwent bariatric surgery in Sweden, 2015. Revision surgery (standardized OR 1.19, 95% confidence interval (CI) 1.14-0.24, p prediction model. Despite high specificity, the sensitivity of the model was low. Revision surgery, high age, low BMI, large waist circumference, and dyspepsia/GERD were associated with an increased risk for severe postoperative complication. The prediction model based on these factors, however, had a sensitivity that was too low to predict risk in the individual patient case.

Complex versus simple models: ion-channel cardiac toxicity prediction.

Science.gov (United States)

Mistry, Hitesh B

2018-01-01

There is growing interest in applying detailed mathematical models of the heart for ion-channel related cardiac toxicity prediction. However, a debate as to whether such complex models are required exists. Here an assessment in the predictive performance between two established large-scale biophysical cardiac models and a simple linear model B net was conducted. Three ion-channel data-sets were extracted from literature. Each compound was designated a cardiac risk category using two different classification schemes based on information within CredibleMeds. The predictive performance of each model within each data-set for each classification scheme was assessed via a leave-one-out cross validation. Overall the B net model performed equally as well as the leading cardiac models in two of the data-sets and outperformed both cardiac models on the latest. These results highlight the importance of benchmarking complex versus simple models but also encourage the development of simple models.

Complex versus simple models: ion-channel cardiac toxicity prediction

Directory of Open Access Journals (Sweden)

Hitesh B. Mistry

2018-02-01

Full Text Available There is growing interest in applying detailed mathematical models of the heart for ion-channel related cardiac toxicity prediction. However, a debate as to whether such complex models are required exists. Here an assessment in the predictive performance between two established large-scale biophysical cardiac models and a simple linear model Bnet was conducted. Three ion-channel data-sets were extracted from literature. Each compound was designated a cardiac risk category using two different classification schemes based on information within CredibleMeds. The predictive performance of each model within each data-set for each classification scheme was assessed via a leave-one-out cross validation. Overall the Bnet model performed equally as well as the leading cardiac models in two of the data-sets and outperformed both cardiac models on the latest. These results highlight the importance of benchmarking complex versus simple models but also encourage the development of simple models.

Key Questions in Building Defect Prediction Models in Practice

Science.gov (United States)

Ramler, Rudolf; Wolfmaier, Klaus; Stauder, Erwin; Kossak, Felix; Natschläger, Thomas

The information about which modules of a future version of a software system are defect-prone is a valuable planning aid for quality managers and testers. Defect prediction promises to indicate these defect-prone modules. However, constructing effective defect prediction models in an industrial setting involves a number of key questions. In this paper we discuss ten key questions identified in context of establishing defect prediction in a large software development project. Seven consecutive versions of the software system have been used to construct and validate defect prediction models for system test planning. Furthermore, the paper presents initial empirical results from the studied project and, by this means, contributes answers to the identified questions.

Predicting Market Impact Costs Using Nonparametric Machine Learning Models.

Science.gov (United States)

Park, Saerom; Lee, Jaewook; Son, Youngdoo

2016-01-01

Market impact cost is the most significant portion of implicit transaction costs that can reduce the overall transaction cost, although it cannot be measured directly. In this paper, we employed the state-of-the-art nonparametric machine learning models: neural networks, Bayesian neural network, Gaussian process, and support vector regression, to predict market impact cost accurately and to provide the predictive model that is versatile in the number of variables. We collected a large amount of real single transaction data of US stock market from Bloomberg Terminal and generated three independent input variables. As a result, most nonparametric machine learning models outperformed a-state-of-the-art benchmark parametric model such as I-star model in four error measures. Although these models encounter certain difficulties in separating the permanent and temporary cost directly, nonparametric machine learning models can be good alternatives in reducing transaction costs by considerably improving in prediction performance.

Predictive Models, How good are they?

DEFF Research Database (Denmark)

Kasch, Helge

The WAD grading system has been used for more than 20 years by now. It has shown long-term viability, but with strengths and limitations. New bio-psychosocial assessment of the acute whiplash injured subject may provide better prediction of long-term disability and pain. Furthermore, the emerging......-up. It is important to obtain prospective identification of the relevant risk underreported disability could, if we were able to expose these hidden “risk-factors” during our consultations, provide us with better predictive models. New data from large clinical studies will present exciting new genetic risk markers...

Risk predictive modelling for diabetes and cardiovascular disease.

Science.gov (United States)

Kengne, Andre Pascal; Masconi, Katya; Mbanya, Vivian Nchanchou; Lekoubou, Alain; Echouffo-Tcheugui, Justin Basile; Matsha, Tandi E

2014-02-01

Absolute risk models or clinical prediction models have been incorporated in guidelines, and are increasingly advocated as tools to assist risk stratification and guide prevention and treatments decisions relating to common health conditions such as cardiovascular disease (CVD) and diabetes mellitus. We have reviewed the historical development and principles of prediction research, including their statistical underpinning, as well as implications for routine practice, with a focus on predictive modelling for CVD and diabetes. Predictive modelling for CVD risk, which has developed over the last five decades, has been largely influenced by the Framingham Heart Study investigators, while it is only ∼20 years ago that similar efforts were started in the field of diabetes. Identification of predictive factors is an important preliminary step which provides the knowledge base on potential predictors to be tested for inclusion during the statistical derivation of the final model. The derived models must then be tested both on the development sample (internal validation) and on other populations in different settings (external validation). Updating procedures (e.g. recalibration) should be used to improve the performance of models that fail the tests of external validation. Ultimately, the effect of introducing validated models in routine practice on the process and outcomes of care as well as its cost-effectiveness should be tested in impact studies before wide dissemination of models beyond the research context. Several predictions models have been developed for CVD or diabetes, but very few have been externally validated or tested in impact studies, and their comparative performance has yet to be fully assessed. A shift of focus from developing new CVD or diabetes prediction models to validating the existing ones will improve their adoption in routine practice.

Computational Techniques for Model Predictive Control of Large-Scale Systems with Continuous-Valued and Discrete-Valued Inputs

Directory of Open Access Journals (Sweden)

Koichi Kobayashi

2013-01-01

Full Text Available We propose computational techniques for model predictive control of large-scale systems with both continuous-valued control inputs and discrete-valued control inputs, which are a class of hybrid systems. In the proposed method, we introduce the notion of virtual control inputs, which are obtained by relaxing discrete-valued control inputs to continuous variables. In online computation, first, we find continuous-valued control inputs and virtual control inputs minimizing a cost function. Next, using the obtained virtual control inputs, only discrete-valued control inputs at the current time are computed in each subsystem. In addition, we also discuss the effect of quantization errors. Finally, the effectiveness of the proposed method is shown by a numerical example. The proposed method enables us to reduce and decentralize the computation load.

Aero-Acoustic Modelling using Large Eddy Simulation

International Nuclear Information System (INIS)

Shen, W Z; Soerensen, J N

2007-01-01

The splitting technique for aero-acoustic computations is extended to simulate three-dimensional flow and acoustic waves from airfoils. The aero-acoustic model is coupled to a sub-grid-scale turbulence model for Large-Eddy Simulations. In the first test case, the model is applied to compute laminar flow past a NACA 0015 airfoil at a Reynolds number of 800, a Mach number of 0.2 and an angle of attack of 20 deg. The model is then applied to compute turbulent flow past a NACA 0015 airfoil at a Reynolds number of 100 000, a Mach number of 0.2 and an angle of attack of 20 deg. The predicted noise spectrum is compared to experimental data

Standardizing the performance evaluation of short-term wind prediction models

DEFF Research Database (Denmark)

Madsen, Henrik; Pinson, Pierre; Kariniotakis, G.

2005-01-01

Short-term wind power prediction is a primary requirement for efficient large-scale integration of wind generation in power systems and electricity markets. The choice of an appropriate prediction model among the numerous available models is not trivial, and has to be based on an objective...... evaluation of model performance. This paper proposes a standardized protocol for the evaluation of short-term wind-poser preciction systems. A number of reference prediction models are also described, and their use for performance comparison is analysed. The use of the protocol is demonstrated using results...... from both on-shore and off-shore wind forms. The work was developed in the frame of the Anemos project (EU R&D project) where the protocol has been used to evaluate more than 10 prediction systems....

Baryogenesis model predicting antimatter in the Universe

International Nuclear Information System (INIS)

Kirilova, D.

2003-01-01

Cosmic ray and gamma-ray data do not rule out antimatter domains in the Universe, separated at distances bigger than 10 Mpc from us. Hence, it is interesting to analyze the possible generation of vast antimatter structures during the early Universe evolution. We discuss a SUSY-condensate baryogenesis model, predicting large separated regions of matter and antimatter. The model provides generation of the small locally observed baryon asymmetry for a natural initial conditions, it predicts vast antimatter domains, separated from the matter ones by baryonically empty voids. The characteristic scale of antimatter regions and their distance from the matter ones is in accordance with observational constraints from cosmic ray, gamma-ray and cosmic microwave background anisotropy data

Predicting Market Impact Costs Using Nonparametric Machine Learning Models.

Directory of Open Access Journals (Sweden)

Saerom Park

Full Text Available Market impact cost is the most significant portion of implicit transaction costs that can reduce the overall transaction cost, although it cannot be measured directly. In this paper, we employed the state-of-the-art nonparametric machine learning models: neural networks, Bayesian neural network, Gaussian process, and support vector regression, to predict market impact cost accurately and to provide the predictive model that is versatile in the number of variables. We collected a large amount of real single transaction data of US stock market from Bloomberg Terminal and generated three independent input variables. As a result, most nonparametric machine learning models outperformed a-state-of-the-art benchmark parametric model such as I-star model in four error measures. Although these models encounter certain difficulties in separating the permanent and temporary cost directly, nonparametric machine learning models can be good alternatives in reducing transaction costs by considerably improving in prediction performance.

Optimization of large-scale heterogeneous system-of-systems models.

Energy Technology Data Exchange (ETDEWEB)

Parekh, Ojas; Watson, Jean-Paul; Phillips, Cynthia Ann; Siirola, John; Swiler, Laura Painton; Hough, Patricia Diane (Sandia National Laboratories, Livermore, CA); Lee, Herbert K. H. (University of California, Santa Cruz, Santa Cruz, CA); Hart, William Eugene; Gray, Genetha Anne (Sandia National Laboratories, Livermore, CA); Woodruff, David L. (University of California, Davis, Davis, CA)

2012-01-01

Decision makers increasingly rely on large-scale computational models to simulate and analyze complex man-made systems. For example, computational models of national infrastructures are being used to inform government policy, assess economic and national security risks, evaluate infrastructure interdependencies, and plan for the growth and evolution of infrastructure capabilities. A major challenge for decision makers is the analysis of national-scale models that are composed of interacting systems: effective integration of system models is difficult, there are many parameters to analyze in these systems, and fundamental modeling uncertainties complicate analysis. This project is developing optimization methods to effectively represent and analyze large-scale heterogeneous system of systems (HSoS) models, which have emerged as a promising approach for describing such complex man-made systems. These optimization methods enable decision makers to predict future system behavior, manage system risk, assess tradeoffs between system criteria, and identify critical modeling uncertainties.

Scaling predictive modeling in drug development with cloud computing.

Science.gov (United States)

Moghadam, Behrooz Torabi; Alvarsson, Jonathan; Holm, Marcus; Eklund, Martin; Carlsson, Lars; Spjuth, Ola

2015-01-26

Growing data sets with increased time for analysis is hampering predictive modeling in drug discovery. Model building can be carried out on high-performance computer clusters, but these can be expensive to purchase and maintain. We have evaluated ligand-based modeling on cloud computing resources where computations are parallelized and run on the Amazon Elastic Cloud. We trained models on open data sets of varying sizes for the end points logP and Ames mutagenicity and compare with model building parallelized on a traditional high-performance computing cluster. We show that while high-performance computing results in faster model building, the use of cloud computing resources is feasible for large data sets and scales well within cloud instances. An additional advantage of cloud computing is that the costs of predictive models can be easily quantified, and a choice can be made between speed and economy. The easy access to computational resources with no up-front investments makes cloud computing an attractive alternative for scientists, especially for those without access to a supercomputer, and our study shows that it enables cost-efficient modeling of large data sets on demand within reasonable time.

Interior Noise Predictions in the Preliminary Design of the Large Civil Tiltrotor (LCTR2)

Science.gov (United States)

Grosveld, Ferdinand W.; Cabell, Randolph H.; Boyd, David D.

2013-01-01

A prediction scheme was established to compute sound pressure levels in the interior of a simplified cabin model of the second generation Large Civil Tiltrotor (LCTR2) during cruise conditions, while being excited by turbulent boundary layer flow over the fuselage, or by tiltrotor blade loading and thickness noise. Finite element models of the cabin structure, interior acoustic space, and acoustically absorbent (poro-elastic) materials in the fuselage were generated and combined into a coupled structural-acoustic model. Fluctuating power spectral densities were computed according to the Efimtsov turbulent boundary layer excitation model. Noise associated with the tiltrotor blades was predicted in the time domain as fluctuating surface pressures and converted to power spectral densities at the fuselage skin finite element nodes. A hybrid finite element (FE) approach was used to compute the low frequency acoustic cabin response over the frequency range 6-141 Hz with a 1 Hz bandwidth, and the Statistical Energy Analysis (SEA) approach was used to predict the interior noise for the 125-8000 Hz one-third octave bands.

Hybrid Reynolds-Averaged/Large Eddy Simulation of a Cavity Flameholder; Assessment of Modeling Sensitivities

Science.gov (United States)

Baurle, R. A.

2015-01-01

Steady-state and scale-resolving simulations have been performed for flow in and around a model scramjet combustor flameholder. The cases simulated corresponded to those used to examine this flowfield experimentally using particle image velocimetry. A variety of turbulence models were used for the steady-state Reynolds-averaged simulations which included both linear and non-linear eddy viscosity models. The scale-resolving simulations used a hybrid Reynolds-averaged / large eddy simulation strategy that is designed to be a large eddy simulation everywhere except in the inner portion (log layer and below) of the boundary layer. Hence, this formulation can be regarded as a wall-modeled large eddy simulation. This effort was undertaken to formally assess the performance of the hybrid Reynolds-averaged / large eddy simulation modeling approach in a flowfield of interest to the scramjet research community. The numerical errors were quantified for both the steady-state and scale-resolving simulations prior to making any claims of predictive accuracy relative to the measurements. The steady-state Reynolds-averaged results showed a high degree of variability when comparing the predictions obtained from each turbulence model, with the non-linear eddy viscosity model (an explicit algebraic stress model) providing the most accurate prediction of the measured values. The hybrid Reynolds-averaged/large eddy simulation results were carefully scrutinized to ensure that even the coarsest grid had an acceptable level of resolution for large eddy simulation, and that the time-averaged statistics were acceptably accurate. The autocorrelation and its Fourier transform were the primary tools used for this assessment. The statistics extracted from the hybrid simulation strategy proved to be more accurate than the Reynolds-averaged results obtained using the linear eddy viscosity models. However, there was no predictive improvement noted over the results obtained from the explicit

Integrating Remote Sensing Information Into A Distributed Hydrological Model for Improving Water Budget Predictions in Large-scale Basins through Data Assimilation

Science.gov (United States)

Qin, Changbo; Jia, Yangwen; Su, Z.(Bob); Zhou, Zuhao; Qiu, Yaqin; Suhui, Shen

2008-01-01

This paper investigates whether remote sensing evapotranspiration estimates can be integrated by means of data assimilation into a distributed hydrological model for improving the predictions of spatial water distribution over a large river basin with an area of 317,800 km2. A series of available MODIS satellite images over the Haihe River basin in China are used for the year 2005. Evapotranspiration is retrieved from these 1×1 km resolution images using the SEBS (Surface Energy Balance System) algorithm. The physically-based distributed model WEP-L (Water and Energy transfer Process in Large river basins) is used to compute the water balance of the Haihe River basin in the same year. Comparison between model-derived and remote sensing retrieval basin-averaged evapotranspiration estimates shows a good piecewise linear relationship, but their spatial distribution within the Haihe basin is different. The remote sensing derived evapotranspiration shows variability at finer scales. An extended Kalman filter (EKF) data assimilation algorithm, suitable for non-linear problems, is used. Assimilation results indicate that remote sensing observations have a potentially important role in providing spatial information to the assimilation system for the spatially optical hydrological parameterization of the model. This is especially important for large basins, such as the Haihe River basin in this study. Combining and integrating the capabilities of and information from model simulation and remote sensing techniques may provide the best spatial and temporal characteristics for hydrological states/fluxes, and would be both appealing and necessary for improving our knowledge of fundamental hydrological processes and for addressing important water resource management problems. PMID:27879946

Prediction Model for Gastric Cancer Incidence in Korean Population.

Directory of Open Access Journals (Sweden)

Bang Wool Eom

Full Text Available Predicting high risk groups for gastric cancer and motivating these groups to receive regular checkups is required for the early detection of gastric cancer. The aim of this study is was to develop a prediction model for gastric cancer incidence based on a large population-based cohort in Korea.Based on the National Health Insurance Corporation data, we analyzed 10 major risk factors for gastric cancer. The Cox proportional hazards model was used to develop gender specific prediction models for gastric cancer development, and the performance of the developed model in terms of discrimination and calibration was also validated using an independent cohort. Discrimination ability was evaluated using Harrell's C-statistics, and the calibration was evaluated using a calibration plot and slope.During a median of 11.4 years of follow-up, 19,465 (1.4% and 5,579 (0.7% newly developed gastric cancer cases were observed among 1,372,424 men and 804,077 women, respectively. The prediction models included age, BMI, family history, meal regularity, salt preference, alcohol consumption, smoking and physical activity for men, and age, BMI, family history, salt preference, alcohol consumption, and smoking for women. This prediction model showed good accuracy and predictability in both the developing and validation cohorts (C-statistics: 0.764 for men, 0.706 for women.In this study, a prediction model for gastric cancer incidence was developed that displayed a good performance.

New tips for structure prediction by comparative modeling

Science.gov (United States)

Rayan, Anwar

2009-01-01

Comparative modelling is utilized to predict the 3-dimensional conformation of a given protein (target) based on its sequence alignment to experimentally determined protein structure (template). The use of such technique is already rewarding and increasingly widespread in biological research and drug development. The accuracy of the predictions as commonly accepted depends on the score of sequence identity of the target protein to the template. To assess the relationship between sequence identity and model quality, we carried out an analysis of a set of 4753 sequence and structure alignments. Throughout this research, the model accuracy was measured by root mean square deviations of Cα atoms of the target-template structures. Surprisingly, the results show that sequence identity of the target protein to the template is not a good descriptor to predict the accuracy of the 3-D structure model. However, in a large number of cases, comparative modelling with lower sequence identity of target to template proteins led to more accurate 3-D structure model. As a consequence of this study, we suggest new tips for improving the quality of omparative models, particularly for models whose target-template sequence identity is below 50%. PMID:19255646

In silico modeling to predict drug-induced phospholipidosis

International Nuclear Information System (INIS)

Choi, Sydney S.; Kim, Jae S.; Valerio, Luis G.; Sadrieh, Nakissa

2013-01-01

Drug-induced phospholipidosis (DIPL) is a preclinical finding during pharmaceutical drug development that has implications on the course of drug development and regulatory safety review. A principal characteristic of drugs inducing DIPL is known to be a cationic amphiphilic structure. This provides evidence for a structure-based explanation and opportunity to analyze properties and structures of drugs with the histopathologic findings for DIPL. In previous work from the FDA, in silico quantitative structure–activity relationship (QSAR) modeling using machine learning approaches has shown promise with a large dataset of drugs but included unconfirmed data as well. In this study, we report the construction and validation of a battery of complementary in silico QSAR models using the FDA's updated database on phospholipidosis, new algorithms and predictive technologies, and in particular, we address high performance with a high-confidence dataset. The results of our modeling for DIPL include rigorous external validation tests showing 80–81% concordance. Furthermore, the predictive performance characteristics include models with high sensitivity and specificity, in most cases above ≥ 80% leading to desired high negative and positive predictivity. These models are intended to be utilized for regulatory toxicology applied science needs in screening new drugs for DIPL. - Highlights: • New in silico models for predicting drug-induced phospholipidosis (DIPL) are described. • The training set data in the models is derived from the FDA's phospholipidosis database. • We find excellent predictivity values of the models based on external validation. • The models can support drug screening and regulatory decision-making on DIPL

CATHARE-2 prediction of large primary to secondary leakage (PRISE) at PSB-VVER experimental facility

Energy Technology Data Exchange (ETDEWEB)

Sabotinov, L.; Chevrier, P. [Institut de Radioprotection et de Surete Nucleaire, 92 - Fontenay aux Roses (France)

2007-07-01

The large primary to secondary leakage (PRISE) is a specific loss-of-coolant accident in VVER reactors, related to the break of the steam generator collector cover, leading to loss of primary mass inventory and possible direct radioactive release to atmosphere. The best estimate thermal-hydraulic computer code CATHARE-2 Version 2.5-1 was used for post-test analysis of a PRISE experiment, conducted at the large scale test facility PSB-VVER in Russia. The PSB rig is 1:300 scaled model of VVER-1000 NPP. The accident is calculated with a 1.4% break size, which corresponds to 100 mm leak from primary to secondary side in the real NPP. A computer model has been developed for CATHARE-2 V2.5-1, taking into account all important components of the PSB facility: reactor model (lower plenum, core, bypass, upper plenum, downcomer), 4 separate loops, pressurizer, horizontal multi-tube steam generators, break section. The secondary side is presented by recirculation model. A large number of sensitivity calculations has been performed regarding break modeling, reactor pressure vessel modeling, counter current flow modeling, hydraulic losses, heat losses, steam generator level regulation. Comparison between calculated and experimental results shows good prediction of the basic thermal-hydraulic phenomena and parameters such as primary and secondary pressures, temperatures, loop flows, etc. Some discrepancies were observed in the calculations of primary mass inventory and loop seal clearance. Nevertheless the final core heat up, which is one of the most important safety criteria, was correctly predicted. (authors)

A survey on computational intelligence approaches for predictive modeling in prostate cancer

OpenAIRE

Cosma, G; Brown, D; Archer, M; Khan, M; Pockley, AG

2017-01-01

Predictive modeling in medicine involves the development of computational models which are capable of analysing large amounts of data in order to predict healthcare outcomes for individual patients. Computational intelligence approaches are suitable when the data to be modelled are too complex forconventional statistical techniques to process quickly and eciently. These advanced approaches are based on mathematical models that have been especially developed for dealing with the uncertainty an...

Homogenization of Large-Scale Movement Models in Ecology

Science.gov (United States)

Garlick, M.J.; Powell, J.A.; Hooten, M.B.; McFarlane, L.R.

2011-01-01

A difficulty in using diffusion models to predict large scale animal population dispersal is that individuals move differently based on local information (as opposed to gradients) in differing habitat types. This can be accommodated by using ecological diffusion. However, real environments are often spatially complex, limiting application of a direct approach. Homogenization for partial differential equations has long been applied to Fickian diffusion (in which average individual movement is organized along gradients of habitat and population density). We derive a homogenization procedure for ecological diffusion and apply it to a simple model for chronic wasting disease in mule deer. Homogenization allows us to determine the impact of small scale (10-100 m) habitat variability on large scale (10-100 km) movement. The procedure generates asymptotic equations for solutions on the large scale with parameters defined by small-scale variation. The simplicity of this homogenization procedure is striking when compared to the multi-dimensional homogenization procedure for Fickian diffusion,and the method will be equally straightforward for more complex models. ?? 2010 Society for Mathematical Biology.

A predictive model of muscle excitations based on muscle modularity for a large repertoire of human locomotion conditions

Directory of Open Access Journals (Sweden)

Jose eGonzalez-Vargas

2015-09-01

Full Text Available Humans can efficiently walk across a large variety of terrains and locomotion conditions with little or no mental effort. It has been hypothesized that the nervous system simplifies neuromuscular control by using muscle synergies, thus organizing multi-muscle activity into a small number of coordinative co-activation modules. In the present study we investigated how muscle modularity is structured across a large repertoire of locomotion conditions including five different speeds and five different ground elevations. For this we have used the non-negative matrix factorization technique in order to explain EMG experimental data with a low-dimensional set of four motor components. In this context each motor components is composed of a non-negative factor and the associated muscle weightings. Furthermore, we have investigated if the proposed descriptive analysis of muscle modularity could be translated into a predictive model that could: 1 Estimate how motor components modulate across locomotion speeds and ground elevations. This implies not only estimating the non-negative factors temporal characteristics, but also the associated muscle weighting variations. 2 Estimate how the resulting muscle excitations modulate across novel locomotion conditions and subjects.The results showed three major distinctive features of muscle modularity: 1 the number of motor components was preserved across all locomotion conditions, 2 the non-negative factors were consistent in shape and timing across all locomotion conditions, and 3 the muscle weightings were modulated as distinctive functions of locomotion speed and ground elevation. Results also showed that the developed predictive model was able to reproduce well the muscle modularity of un-modeled data, i.e. novel subjects and conditions. Muscle weightings were reconstructed with a cross-correlation factor greater than 70% and a root mean square error less than 0.10. Furthermore, the generated muscle excitations

Modeling, Prediction, and Control of Heating Temperature for Tube Billet

Directory of Open Access Journals (Sweden)

Yachun Mao

2015-01-01

Full Text Available Annular furnaces have multivariate, nonlinear, large time lag, and cross coupling characteristics. The prediction and control of the exit temperature of a tube billet are important but difficult. We establish a prediction model for the final temperature of a tube billet through OS-ELM-DRPLS method. We address the complex production characteristics, integrate the advantages of PLS and ELM algorithms in establishing linear and nonlinear models, and consider model update and data lag. Based on the proposed model, we design a prediction control algorithm for tube billet temperature. The algorithm is validated using the practical production data of Baosteel Co., Ltd. Results show that the model achieves the precision required in industrial applications. The temperature of the tube billet can be controlled within the required temperature range through compensation control method.

Impact of modellers' decisions on hydrological a priori predictions

Science.gov (United States)

Holländer, H. M.; Bormann, H.; Blume, T.; Buytaert, W.; Chirico, G. B.; Exbrayat, J.-F.; Gustafsson, D.; Hölzel, H.; Krauße, T.; Kraft, P.; Stoll, S.; Blöschl, G.; Flühler, H.

2014-06-01

In practice, the catchment hydrologist is often confronted with the task of predicting discharge without having the needed records for calibration. Here, we report the discharge predictions of 10 modellers - using the model of their choice - for the man-made Chicken Creek catchment (6 ha, northeast Germany, Gerwin et al., 2009b) and we analyse how well they improved their prediction in three steps based on adding information prior to each following step. The modellers predicted the catchment's hydrological response in its initial phase without having access to the observed records. They used conceptually different physically based models and their modelling experience differed largely. Hence, they encountered two problems: (i) to simulate discharge for an ungauged catchment and (ii) using models that were developed for catchments, which are not in a state of landscape transformation. The prediction exercise was organized in three steps: (1) for the first prediction the modellers received a basic data set describing the catchment to a degree somewhat more complete than usually available for a priori predictions of ungauged catchments; they did not obtain information on stream flow, soil moisture, nor groundwater response and had therefore to guess the initial conditions; (2) before the second prediction they inspected the catchment on-site and discussed their first prediction attempt; (3) for their third prediction they were offered additional data by charging them pro forma with the costs for obtaining this additional information. Holländer et al. (2009) discussed the range of predictions obtained in step (1). Here, we detail the modeller's assumptions and decisions in accounting for the various processes. We document the prediction progress as well as the learning process resulting from the availability of added information. For the second and third steps, the progress in prediction quality is evaluated in relation to individual modelling experience and costs of

Modelling bankruptcy prediction models in Slovak companies

Directory of Open Access Journals (Sweden)

Kovacova Maria

2017-01-01

Full Text Available An intensive research from academics and practitioners has been provided regarding models for bankruptcy prediction and credit risk management. In spite of numerous researches focusing on forecasting bankruptcy using traditional statistics techniques (e.g. discriminant analysis and logistic regression and early artificial intelligence models (e.g. artificial neural networks, there is a trend for transition to machine learning models (support vector machines, bagging, boosting, and random forest to predict bankruptcy one year prior to the event. Comparing the performance of this with unconventional approach with results obtained by discriminant analysis, logistic regression, and neural networks application, it has been found that bagging, boosting, and random forest models outperform the others techniques, and that all prediction accuracy in the testing sample improves when the additional variables are included. On the other side the prediction accuracy of old and well known bankruptcy prediction models is quiet high. Therefore, we aim to analyse these in some way old models on the dataset of Slovak companies to validate their prediction ability in specific conditions. Furthermore, these models will be modelled according to new trends by calculating the influence of elimination of selected variables on the overall prediction ability of these models.

Large Animal Stroke Models vs. Rodent Stroke Models, Pros and Cons, and Combination?

Science.gov (United States)

Cai, Bin; Wang, Ning

2016-01-01

Stroke is a leading cause of serious long-term disability worldwide and the second leading cause of death in many countries. Long-time attempts to salvage dying neurons via various neuroprotective agents have failed in stroke translational research, owing in part to the huge gap between animal stroke models and stroke patients, which also suggests that rodent models have limited predictive value and that alternate large animal models are likely to become important in future translational research. The genetic background, physiological characteristics, behavioral characteristics, and brain structure of large animals, especially nonhuman primates, are analogous to humans, and resemble humans in stroke. Moreover, relatively new regional imaging techniques, measurements of regional cerebral blood flow, and sophisticated physiological monitoring can be more easily performed on the same animal at multiple time points. As a result, we can use large animal stroke models to decrease the gap and promote translation of basic science stroke research. At the same time, we should not neglect the disadvantages of the large animal stroke model such as the significant expense and ethical considerations, which can be overcome by rodent models. Rodents should be selected as stroke models for initial testing and primates or cats are desirable as a second species, which was recommended by the Stroke Therapy Academic Industry Roundtable (STAIR) group in 2009.

Selecting Optimal Random Forest Predictive Models: A Case Study on Predicting the Spatial Distribution of Seabed Hardness

Science.gov (United States)

Li, Jin; Tran, Maggie; Siwabessy, Justy

2016-01-01

Spatially continuous predictions of seabed hardness are important baseline environmental information for sustainable management of Australia’s marine jurisdiction. Seabed hardness is often inferred from multibeam backscatter data with unknown accuracy and can be inferred from underwater video footage at limited locations. In this study, we classified the seabed into four classes based on two new seabed hardness classification schemes (i.e., hard90 and hard70). We developed optimal predictive models to predict seabed hardness using random forest (RF) based on the point data of hardness classes and spatially continuous multibeam data. Five feature selection (FS) methods that are variable importance (VI), averaged variable importance (AVI), knowledge informed AVI (KIAVI), Boruta and regularized RF (RRF) were tested based on predictive accuracy. Effects of highly correlated, important and unimportant predictors on the accuracy of RF predictive models were examined. Finally, spatial predictions generated using the most accurate models were visually examined and analysed. This study confirmed that: 1) hard90 and hard70 are effective seabed hardness classification schemes; 2) seabed hardness of four classes can be predicted with a high degree of accuracy; 3) the typical approach used to pre-select predictive variables by excluding highly correlated variables needs to be re-examined; 4) the identification of the important and unimportant predictors provides useful guidelines for further improving predictive models; 5) FS methods select the most accurate predictive model(s) instead of the most parsimonious ones, and AVI and Boruta are recommended for future studies; and 6) RF is an effective modelling method with high predictive accuracy for multi-level categorical data and can be applied to ‘small p and large n’ problems in environmental sciences. Additionally, automated computational programs for AVI need to be developed to increase its computational efficiency and

Prediction of monthly rainfall on homogeneous monsoon regions of India based on large scale circulation patterns using Genetic Programming

Science.gov (United States)

Kashid, Satishkumar S.; Maity, Rajib

2012-08-01

SummaryPrediction of Indian Summer Monsoon Rainfall (ISMR) is of vital importance for Indian economy, and it has been remained a great challenge for hydro-meteorologists due to inherent complexities in the climatic systems. The Large-scale atmospheric circulation patterns from tropical Pacific Ocean (ENSO) and those from tropical Indian Ocean (EQUINOO) are established to influence the Indian Summer Monsoon Rainfall. The information of these two large scale atmospheric circulation patterns in terms of their indices is used to model the complex relationship between Indian Summer Monsoon Rainfall and the ENSO as well as EQUINOO indices. However, extracting the signal from such large-scale indices for modeling such complex systems is significantly difficult. Rainfall predictions have been done for 'All India' as one unit, as well as for five 'homogeneous monsoon regions of India', defined by Indian Institute of Tropical Meteorology. Recent 'Artificial Intelligence' tool 'Genetic Programming' (GP) has been employed for modeling such problem. The Genetic Programming approach is found to capture the complex relationship between the monthly Indian Summer Monsoon Rainfall and large scale atmospheric circulation pattern indices - ENSO and EQUINOO. Research findings of this study indicate that GP-derived monthly rainfall forecasting models, that use large-scale atmospheric circulation information are successful in prediction of All India Summer Monsoon Rainfall with correlation coefficient as good as 0.866, which may appears attractive for such a complex system. A separate analysis is carried out for All India Summer Monsoon rainfall for India as one unit, and five homogeneous monsoon regions, based on ENSO and EQUINOO indices of months of March, April and May only, performed at end of month of May. In this case, All India Summer Monsoon Rainfall could be predicted with 0.70 as correlation coefficient with somewhat lesser Correlation Coefficient (C.C.) values for different

A refined regional modeling approach for the Corn Belt - Experiences and recommendations for large-scale integrated modeling

Science.gov (United States)

Panagopoulos, Yiannis; Gassman, Philip W.; Jha, Manoj K.; Kling, Catherine L.; Campbell, Todd; Srinivasan, Raghavan; White, Michael; Arnold, Jeffrey G.

2015-05-01

Nonpoint source pollution from agriculture is the main source of nitrogen and phosphorus in the stream systems of the Corn Belt region in the Midwestern US. This region is comprised of two large river basins, the intensely row-cropped Upper Mississippi River Basin (UMRB) and Ohio-Tennessee River Basin (OTRB), which are considered the key contributing areas for the Northern Gulf of Mexico hypoxic zone according to the US Environmental Protection Agency. Thus, in this area it is of utmost importance to ensure that intensive agriculture for food, feed and biofuel production can coexist with a healthy water environment. To address these objectives within a river basin management context, an integrated modeling system has been constructed with the hydrologic Soil and Water Assessment Tool (SWAT) model, capable of estimating river basin responses to alternative cropping and/or management strategies. To improve modeling performance compared to previous studies and provide a spatially detailed basis for scenario development, this SWAT Corn Belt application incorporates a greatly refined subwatershed structure based on 12-digit hydrologic units or 'subwatersheds' as defined by the US Geological Service. The model setup, calibration and validation are time-demanding and challenging tasks for these large systems, given the scale intensive data requirements, and the need to ensure the reliability of flow and pollutant load predictions at multiple locations. Thus, the objectives of this study are both to comprehensively describe this large-scale modeling approach, providing estimates of pollution and crop production in the region as well as to present strengths and weaknesses of integrated modeling at such a large scale along with how it can be improved on the basis of the current modeling structure and results. The predictions were based on a semi-automatic hydrologic calibration approach for large-scale and spatially detailed modeling studies, with the use of the Sequential

Predictive modeling of coupled multi-physics systems: II. Illustrative application to reactor physics

International Nuclear Information System (INIS)

Cacuci, Dan Gabriel; Badea, Madalina Corina

2014-01-01

Highlights: • We applied the PMCMPS methodology to a paradigm neutron diffusion model. • We underscore the main steps in applying PMCMPS to treat very large coupled systems. • PMCMPS reduces the uncertainties in the optimally predicted responses and model parameters. • PMCMPS is for sequentially treating coupled systems that cannot be treated simultaneously. - Abstract: This work presents paradigm applications to reactor physics of the innovative mathematical methodology for “predictive modeling of coupled multi-physics systems (PMCMPS)” developed by Cacuci (2014). This methodology enables the assimilation of experimental and computational information and computes optimally predicted responses and model parameters with reduced predicted uncertainties, taking fully into account the coupling terms between the multi-physics systems, but using only the computational resources that would be needed to perform predictive modeling on each system separately. The paradigm examples presented in this work are based on a simple neutron diffusion model, chosen so as to enable closed-form solutions with clear physical interpretations. These paradigm examples also illustrate the computational efficiency of the PMCMPS, which enables the assimilation of additional experimental information, with a minimal increase in computational resources, to reduce the uncertainties in predicted responses and best-estimate values for uncertain model parameters, thus illustrating how very large systems can be treated without loss of information in a sequential rather than simultaneous manner

Verification and improvement of a predictive model for radionuclide migration

International Nuclear Information System (INIS)

Miller, C.W.; Benson, L.V.; Carnahan, C.L.

1982-01-01

Prediction of the rates of migration of contaminant chemical species in groundwater flowing through toxic waste repositories is essential to the assessment of a repository's capability of meeting standards for release rates. A large number of chemical transport models, of varying degrees of complexity, have been devised for the purpose of providing this predictive capability. In general, the transport of dissolved chemical species through a water-saturated porous medium is influenced by convection, diffusion/dispersion, sorption, formation of complexes in the aqueous phase, and chemical precipitation. The reliability of predictions made with the models which omit certain of these processes is difficult to assess. A numerical model, CHEMTRN, has been developed to determine which chemical processes govern radionuclide migration. CHEMTRN builds on a model called MCCTM developed previously by Lichtner and Benson

Questioning the Faith - Models and Prediction in Stream Restoration (Invited)

Science.gov (United States)

Wilcock, P.

2013-12-01

River management and restoration demand prediction at and beyond our present ability. Management questions, framed appropriately, can motivate fundamental advances in science, although the connection between research and application is not always easy, useful, or robust. Why is that? This presentation considers the connection between models and management, a connection that requires critical and creative thought on both sides. Essential challenges for managers include clearly defining project objectives and accommodating uncertainty in any model prediction. Essential challenges for the research community include matching the appropriate model to project duration, space, funding, information, and social constraints and clearly presenting answers that are actually useful to managers. Better models do not lead to better management decisions or better designs if the predictions are not relevant to and accepted by managers. In fact, any prediction may be irrelevant if the need for prediction is not recognized. The predictive target must be developed in an active dialog between managers and modelers. This relationship, like any other, can take time to develop. For example, large segments of stream restoration practice have remained resistant to models and prediction because the foundational tenet - that channels built to a certain template will be able to transport the supplied sediment with the available flow - has no essential physical connection between cause and effect. Stream restoration practice can be steered in a predictive direction in which project objectives are defined as predictable attributes and testable hypotheses. If stream restoration design is defined in terms of the desired performance of the channel (static or dynamic, sediment surplus or deficit), then channel properties that provide these attributes can be predicted and a basis exists for testing approximations, models, and predictions.

Speedup predictions on large scientific parallel programs

International Nuclear Information System (INIS)

Williams, E.; Bobrowicz, F.

1985-01-01

How much speedup can we expect for large scientific parallel programs running on supercomputers. For insight into this problem we extend the parallel processing environment currently existing on the Cray X-MP (a shared memory multiprocessor with at most four processors) to a simulated N-processor environment, where N greater than or equal to 1. Several large scientific parallel programs from Los Alamos National Laboratory were run in this simulated environment, and speedups were predicted. A speedup of 14.4 on 16 processors was measured for one of the three most used codes at the Laboratory

Evaluating predictive models of software quality

International Nuclear Information System (INIS)

Ciaschini, V; Canaparo, M; Ronchieri, E; Salomoni, D

2014-01-01

Applications from High Energy Physics scientific community are constantly growing and implemented by a large number of developers. This implies a strong churn on the code and an associated risk of faults, which is unavoidable as long as the software undergoes active evolution. However, the necessities of production systems run counter to this. Stability and predictability are of paramount importance; in addition, a short turn-around time for the defect discovery-correction-deployment cycle is required. A way to reconcile these opposite foci is to use a software quality model to obtain an approximation of the risk before releasing a program to only deliver software with a risk lower than an agreed threshold. In this article we evaluated two quality predictive models to identify the operational risk and the quality of some software products. We applied these models to the development history of several EMI packages with intent to discover the risk factor of each product and compare it with its real history. We attempted to determine if the models reasonably maps reality for the applications under evaluation, and finally we concluded suggesting directions for further studies.

Evaluating Predictive Models of Software Quality

Science.gov (United States)

Ciaschini, V.; Canaparo, M.; Ronchieri, E.; Salomoni, D.

2014-06-01

Applications from High Energy Physics scientific community are constantly growing and implemented by a large number of developers. This implies a strong churn on the code and an associated risk of faults, which is unavoidable as long as the software undergoes active evolution. However, the necessities of production systems run counter to this. Stability and predictability are of paramount importance; in addition, a short turn-around time for the defect discovery-correction-deployment cycle is required. A way to reconcile these opposite foci is to use a software quality model to obtain an approximation of the risk before releasing a program to only deliver software with a risk lower than an agreed threshold. In this article we evaluated two quality predictive models to identify the operational risk and the quality of some software products. We applied these models to the development history of several EMI packages with intent to discover the risk factor of each product and compare it with its real history. We attempted to determine if the models reasonably maps reality for the applications under evaluation, and finally we concluded suggesting directions for further studies.

Flash flood prediction in large dams using neural networks

Science.gov (United States)

Múnera Estrada, J. C.; García Bartual, R.

2009-04-01

A flow forecasting methodology is presented as a support tool for flood management in large dams. The practical and efficient use of hydrological real-time measurements is necessary to operate early warning systems for flood disasters prevention, either in natural catchments or in those regulated with reservoirs. In this latter case, the optimal dam operation during flood scenarios should reduce the downstream risks, and at the same time achieve a compromise between different goals: structural security, minimize predictions uncertainty and water resources system management objectives. Downstream constraints depend basically on the geomorphology of the valley, the critical flow thresholds for flooding, the land use and vulnerability associated with human settlements and their economic activities. A dam operation during a flood event thus requires appropriate strategies depending on the flood magnitude and the initial freeboard at the reservoir. The most important difficulty arises from the inherently stochastic character of peak rainfall intensities, their strong spatial and temporal variability, and the highly nonlinear response of semiarid catchments resulting from initial soil moisture condition and the dominant flow mechanisms. The practical integration of a flow prediction model in a real-time system should include combined techniques of pre-processing, data verification and completion, assimilation of information and implementation of real time filters depending on the system characteristics. This work explores the behaviour of real-time flood forecast algorithms based on artificial neural networks (ANN) techniques, in the River Meca catchment (Huelva, Spain), regulated by El Sancho dam. The dam is equipped with three Taintor gates of 12x6 meters. The hydrological data network includes five high-resolution automatic pluviometers (dt=10 min) and three high precision water level sensors in the reservoir. A cross correlation analysis between precipitation data

Evaluation of Presumed Probability-Density-Function Models in Non-Premixed Flames by using Large Eddy Simulation

International Nuclear Information System (INIS)

Cao Hong-Jun; Zhang Hui-Qiang; Lin Wen-Yi

2012-01-01

Four kinds of presumed probability-density-function (PDF) models for non-premixed turbulent combustion are evaluated in flames with various stoichiometric mixture fractions by using large eddy simulation (LES). The LES code is validated by the experimental data of a classical turbulent jet flame (Sandia flame D). The mean and rms temperatures obtained by the presumed PDF models are compared with the LES results. The β-function model achieves a good prediction for different flames. The predicted rms temperature by using the double-δ function model is very small and unphysical in the vicinity of the maximum mean temperature. The clip-Gaussian model and the multi-δ function model make a worse prediction of the extremely fuel-rich or fuel-lean side due to the clip at the boundary of the mixture fraction space. The results also show that the overall prediction performance of presumed PDF models is better at mediate stoichiometric mixture fractions than that at very small or very large ones. (fundamental areas of phenomenology(including applications))

Lumped hydrological models is an Occam' razor for runoff modeling in large Russian Arctic basins

OpenAIRE

Ayzel Georgy

2018-01-01

This study is aimed to investigate the possibility of three lumped hydrological models to predict daily runoff of large-scale Arctic basins for the modern period (1979-2014) in the case of substantial data scarcity. All models were driven only by meteorological forcing reanalysis dataset without any additional information about landscape, soil or vegetation cover properties of studied basins. We found limitations of model parameters calibration in ungauged basins using global optimization alg...

Quantifying the predictive consequences of model error with linear subspace analysis

Science.gov (United States)

White, Jeremy T.; Doherty, John E.; Hughes, Joseph D.

2014-01-01

All computer models are simplified and imperfect simulators of complex natural systems. The discrepancy arising from simplification induces bias in model predictions, which may be amplified by the process of model calibration. This paper presents a new method to identify and quantify the predictive consequences of calibrating a simplified computer model. The method is based on linear theory, and it scales efficiently to the large numbers of parameters and observations characteristic of groundwater and petroleum reservoir models. The method is applied to a range of predictions made with a synthetic integrated surface-water/groundwater model with thousands of parameters. Several different observation processing strategies and parameterization/regularization approaches are examined in detail, including use of the Karhunen-Loève parameter transformation. Predictive bias arising from model error is shown to be prediction specific and often invisible to the modeler. The amount of calibration-induced bias is influenced by several factors, including how expert knowledge is applied in the design of parameterization schemes, the number of parameters adjusted during calibration, how observations and model-generated counterparts are processed, and the level of fit with observations achieved through calibration. Failure to properly implement any of these factors in a prediction-specific manner may increase the potential for predictive bias in ways that are not visible to the calibration and uncertainty analysis process.

Simplest simulation model for three-dimensional xenon oscillations in large PWRs

International Nuclear Information System (INIS)

Shimazu, Yoichiro

2004-01-01

Xenon oscillations in large PWRs are well understood and there have been no operational problems remained. However, in order to suppress the oscillations effectively, optimal control strategy is preferable. Generally speaking in such optimality search based on the modern control theory, a large volume of transient core analyses is required. For example, three dimensional core calculations are inevitable for the analyses of radial oscillations. From this point of view, a very simple 3-D model is proposed, which is based on a reactor model of only four points. As in the actual reactor operation, the magnitude of xenon oscillations should be limited from the view point of safety, the model further assumes that the neutron leakage can be also small or even constant. It can explicitly use reactor parameters such as reactivity coefficients and control rod worth directly. The model is so simplified as described above that it can predict oscillation behavior in a very short calculation time even on a PC. However the prediction result is good. The validity of the model in comparison with measured data and the applications are discussed. (author)

Coupling a Mesoscale Numerical Weather Prediction Model with Large-Eddy Simulation for Realistic Wind Plant Aerodynamics Simulations (Poster)

Energy Technology Data Exchange (ETDEWEB)

Draxl, C.; Churchfield, M.; Mirocha, J.; Lee, S.; Lundquist, J.; Michalakes, J.; Moriarty, P.; Purkayastha, A.; Sprague, M.; Vanderwende, B.

2014-06-01

Wind plant aerodynamics are influenced by a combination of microscale and mesoscale phenomena. Incorporating mesoscale atmospheric forcing (e.g., diurnal cycles and frontal passages) into wind plant simulations can lead to a more accurate representation of microscale flows, aerodynamics, and wind turbine/plant performance. Our goal is to couple a numerical weather prediction model that can represent mesoscale flow [specifically the Weather Research and Forecasting model] with a microscale LES model (OpenFOAM) that can predict microscale turbulence and wake losses.

Solving large linear systems in an implicit thermohaline ocean model

NARCIS (Netherlands)

de Niet, Arie Christiaan

2007-01-01

The climate on earth is largely determined by the global ocean circulation. Hence it is important to predict how the flow will react to perturbation by for example melting icecaps. To answer questions about the stability of the global ocean flow, a computer model has been developed that is able to

Predictive modeling of complications.

Science.gov (United States)

Osorio, Joseph A; Scheer, Justin K; Ames, Christopher P

2016-09-01

Predictive analytic algorithms are designed to identify patterns in the data that allow for accurate predictions without the need for a hypothesis. Therefore, predictive modeling can provide detailed and patient-specific information that can be readily applied when discussing the risks of surgery with a patient. There are few studies using predictive modeling techniques in the adult spine surgery literature. These types of studies represent the beginning of the use of predictive analytics in spine surgery outcomes. We will discuss the advancements in the field of spine surgery with respect to predictive analytics, the controversies surrounding the technique, and the future directions.

The North American Multi-Model Ensemble (NMME): Phase-1 Seasonal to Interannual Prediction, Phase-2 Toward Developing Intra-Seasonal Prediction

Science.gov (United States)

Kirtman, Ben P.; Min, Dughong; Infanti, Johnna M.; Kinter, James L., III; Paolino, Daniel A.; Zhang, Qin; vandenDool, Huug; Saha, Suranjana; Mendez, Malaquias Pena; Becker, Emily;

Application of a predictive Bayesian model to environmental accounting.

Science.gov (United States)

Anex, R P; Englehardt, J D

2001-03-30

Environmental accounting techniques are intended to capture important environmental costs and benefits that are often overlooked in standard accounting practices. Environmental accounting methods themselves often ignore or inadequately represent large but highly uncertain environmental costs and costs conditioned by specific prior events. Use of a predictive Bayesian model is demonstrated for the assessment of such highly uncertain environmental and contingent costs. The predictive Bayesian approach presented generates probability distributions for the quantity of interest (rather than parameters thereof). A spreadsheet implementation of a previously proposed predictive Bayesian model, extended to represent contingent costs, is described and used to evaluate whether a firm should undertake an accelerated phase-out of its PCB containing transformers. Variability and uncertainty (due to lack of information) in transformer accident frequency and severity are assessed simultaneously using a combination of historical accident data, engineering model-based cost estimates, and subjective judgement. Model results are compared using several different risk measures. Use of the model for incorporation of environmental risk management into a company's overall risk management strategy is discussed.

Uncertainties in predicting rice yield by current crop models under a wide range of climatic conditions

NARCIS (Netherlands)

Li, T.; Hasegawa, T.; Yin, X.; Zhu, Y.; Boote, K.; Adam, M.; Bregaglio, S.; Buis, S.; Confalonieri, R.; Fumoto, T.; Gaydon, D.; Marcaida III, M.; Nakagawa, H.; Oriol, P.; Ruane, A.C.; Ruget, F.; Singh, B.; Singh, U.; Tang, L.; Yoshida, H.; Zhang, Z.; Bouman, B.

2015-01-01

Predicting rice (Oryza sativa) productivity under future climates is important for global food security. Ecophysiological crop models in combination with climate model outputs are commonly used in yield prediction, but uncertainties associated with crop models remain largely unquantified. We

Detailed physical properties prediction of pure methyl esters for biodiesel combustion modeling

International Nuclear Information System (INIS)

An, H.; Yang, W.M.; Maghbouli, A.; Chou, S.K.; Chua, K.J.

2013-01-01

Highlights: ► Group contribution methods from molecular level have been used for the prediction. ► Complete prediction of the physical properties for 5 methyl esters has been done. ► The predicted results can be very useful for biodiesel combustion modeling. ► Various models have been compared and the best model has been identified. ► Predicted properties are over large temperature ranges with excellent accuracies. -- Abstract: In order to accurately simulate the fuel spray, atomization, combustion and emission formation processes of a diesel engine fueled with biodiesel, adequate knowledge of biodiesel’s physical properties is desired. The objective of this work is to do a detailed physical properties prediction for the five major methyl esters of biodiesel for combustion modeling. The physical properties considered in this study are: normal boiling point, critical properties, vapor pressure, and latent heat of vaporization, liquid density, liquid viscosity, liquid thermal conductivity, gas diffusion coefficients and surface tension. For each physical property, the best prediction model has been identified, and very good agreements have been obtained between the predicted results and the published data where available. The calculated results can be used as key references for biodiesel combustion modeling.

Application of Large-Scale Database-Based Online Modeling to Plant State Long-Term Estimation

Science.gov (United States)

Ogawa, Masatoshi; Ogai, Harutoshi

Recently, attention has been drawn to the local modeling techniques of a new idea called “Just-In-Time (JIT) modeling”. To apply “JIT modeling” to a large amount of database online, “Large-scale database-based Online Modeling (LOM)” has been proposed. LOM is a technique that makes the retrieval of neighboring data more efficient by using both “stepwise selection” and quantization. In order to predict the long-term state of the plant without using future data of manipulated variables, an Extended Sequential Prediction method of LOM (ESP-LOM) has been proposed. In this paper, the LOM and the ESP-LOM are introduced.

Why choose Random Forest to predict rare species distribution with few samples in large undersampled areas? Three Asian crane species models provide supporting evidence

Directory of Open Access Journals (Sweden)

Chunrong Mi

2017-01-01

Full Text Available Species distribution models (SDMs have become an essential tool in ecology, biogeography, evolution and, more recently, in conservation biology. How to generalize species distributions in large undersampled areas, especially with few samples, is a fundamental issue of SDMs. In order to explore this issue, we used the best available presence records for the Hooded Crane (Grus monacha, n = 33, White-naped Crane (Grus vipio, n = 40, and Black-necked Crane (Grus nigricollis, n = 75 in China as three case studies, employing four powerful and commonly used machine learning algorithms to map the breeding distributions of the three species: TreeNet (Stochastic Gradient Boosting, Boosted Regression Tree Model, Random Forest, CART (Classification and Regression Tree and Maxent (Maximum Entropy Models. In addition, we developed an ensemble forecast by averaging predicted probability of the above four models results. Commonly used model performance metrics (Area under ROC (AUC and true skill statistic (TSS were employed to evaluate model accuracy. The latest satellite tracking data and compiled literature data were used as two independent testing datasets to confront model predictions. We found Random Forest demonstrated the best performance for the most assessment method, provided a better model fit to the testing data, and achieved better species range maps for each crane species in undersampled areas. Random Forest has been generally available for more than 20 years and has been known to perform extremely well in ecological predictions. However, while increasingly on the rise, its potential is still widely underused in conservation, (spatial ecological applications and for inference. Our results show that it informs ecological and biogeographical theories as well as being suitable for conservation applications, specifically when the study area is undersampled. This method helps to save model-selection time and effort, and allows robust and rapid

Why choose Random Forest to predict rare species distribution with few samples in large undersampled areas? Three Asian crane species models provide supporting evidence.

Science.gov (United States)

Mi, Chunrong; Huettmann, Falk; Guo, Yumin; Han, Xuesong; Wen, Lijia

2017-01-01

Species distribution models (SDMs) have become an essential tool in ecology, biogeography, evolution and, more recently, in conservation biology. How to generalize species distributions in large undersampled areas, especially with few samples, is a fundamental issue of SDMs. In order to explore this issue, we used the best available presence records for the Hooded Crane ( Grus monacha , n  = 33), White-naped Crane ( Grus vipio , n  = 40), and Black-necked Crane ( Grus nigricollis , n  = 75) in China as three case studies, employing four powerful and commonly used machine learning algorithms to map the breeding distributions of the three species: TreeNet (Stochastic Gradient Boosting, Boosted Regression Tree Model), Random Forest, CART (Classification and Regression Tree) and Maxent (Maximum Entropy Models). In addition, we developed an ensemble forecast by averaging predicted probability of the above four models results. Commonly used model performance metrics (Area under ROC (AUC) and true skill statistic (TSS)) were employed to evaluate model accuracy. The latest satellite tracking data and compiled literature data were used as two independent testing datasets to confront model predictions. We found Random Forest demonstrated the best performance for the most assessment method, provided a better model fit to the testing data, and achieved better species range maps for each crane species in undersampled areas. Random Forest has been generally available for more than 20 years and has been known to perform extremely well in ecological predictions. However, while increasingly on the rise, its potential is still widely underused in conservation, (spatial) ecological applications and for inference. Our results show that it informs ecological and biogeographical theories as well as being suitable for conservation applications, specifically when the study area is undersampled. This method helps to save model-selection time and effort, and allows robust and rapid

Stochastic backscatter modelling for the prediction of pollutant removal from an urban street canyon: A large-eddy simulation

Science.gov (United States)

O'Neill, J. J.; Cai, X.-M.; Kinnersley, R.

2016-10-01

The large-eddy simulation (LES) approach has recently exhibited its appealing capability of capturing turbulent processes inside street canyons and the urban boundary layer aloft, and its potential for deriving the bulk parameters adopted in low-cost operational urban dispersion models. However, the thin roof-level shear layer may be under-resolved in most LES set-ups and thus sophisticated subgrid-scale (SGS) parameterisations may be required. In this paper, we consider the important case of pollutant removal from an urban street canyon of unit aspect ratio (i.e. building height equal to street width) with the external flow perpendicular to the street. We show that by employing a stochastic SGS model that explicitly accounts for backscatter (energy transfer from unresolved to resolved scales), the pollutant removal process is better simulated compared with the use of a simpler (fully dissipative) but widely-used SGS model. The backscatter induces additional mixing within the shear layer which acts to increase the rate of pollutant removal from the street canyon, giving better agreement with a recent wind-tunnel experiment. The exchange velocity, an important parameter in many operational models that determines the mass transfer between the urban canopy and the external flow, is predicted to be around 15% larger with the backscatter SGS model; consequently, the steady-state mean pollutant concentration within the street canyon is around 15% lower. A database of exchange velocities for various other urban configurations could be generated and used as improved input for operational street canyon models.

Candidate Prediction Models and Methods

DEFF Research Database (Denmark)

Nielsen, Henrik Aalborg; Nielsen, Torben Skov; Madsen, Henrik

2005-01-01

This document lists candidate prediction models for Work Package 3 (WP3) of the PSO-project called ``Intelligent wind power prediction systems'' (FU4101). The main focus is on the models transforming numerical weather predictions into predictions of power production. The document also outlines...... the possibilities w.r.t. different numerical weather predictions actually available to the project....

Generalized additive models used to predict species abundance in the Gulf of Mexico: an ecosystem modeling tool.

Directory of Open Access Journals (Sweden)

Michael Drexler

Full Text Available Spatially explicit ecosystem models of all types require an initial allocation of biomass, often in areas where fisheries independent abundance estimates do not exist. A generalized additive modelling (GAM approach is used to describe the abundance of 40 species groups (i.e. functional groups across the Gulf of Mexico (GoM using a large fisheries independent data set (SEAMAP and climate scale oceanographic conditions. Predictor variables included in the model are chlorophyll a, sediment type, dissolved oxygen, temperature, and depth. Despite the presence of a large number of zeros in the data, a single GAM using a negative binomial distribution was suitable to make predictions of abundance for multiple functional groups. We present an example case study using pink shrimp (Farfantepenaeus duroarum and compare the results to known distributions. The model successfully predicts the known areas of high abundance in the GoM, including those areas where no data was inputted into the model fitting. Overall, the model reliably captures areas of high and low abundance for the large majority of functional groups observed in SEAMAP. The result of this method allows for the objective setting of spatial distributions for numerous functional groups across a modeling domain, even where abundance data may not exist.

Data Quality Enhanced Prediction Model for Massive Plant Data

International Nuclear Information System (INIS)

Park, Moon-Ghu; Kang, Seong-Ki; Shin, Hajin

2016-01-01

This paper introduces an integrated signal preconditioning and model prediction mainly by kernel functions. The performance and benefits of the methods are demonstrated by a case study with measurement data from a power plant and its components transient data. The developed methods will be applied as a part of monitoring massive or big data platform where human experts cannot detect the fault behaviors due to too large size of the measurements. Recent extensive efforts for on-line monitoring implementation insists that a big surprise in the modeling for predicting process variables was the extent of data quality problems in measurement data especially for data-driven modeling. Bad data for training will be learned as normal and can make significant degrade in prediction performance. For this reason, the quantity and quality of measurement data in modeling phase need special care. Bad quality data must be removed from training sets to the bad data considered as normal system behavior. This paper presented an integrated structure of supervisory system for monitoring the plants or sensors performance. The quality of the data-driven model is improved with a bilateral kernel filter for preprocessing of the noisy data. The prediction module is also based on kernel regression having the same basis with noise filter. The model structure is optimized by a grouping process with nonlinear Hoeffding correlation function

Data Quality Enhanced Prediction Model for Massive Plant Data

Energy Technology Data Exchange (ETDEWEB)

Park, Moon-Ghu [Nuclear Engr. Sejong Univ., Seoul (Korea, Republic of); Kang, Seong-Ki [Monitoring and Diagnosis, Suwon (Korea, Republic of); Shin, Hajin [Saint Paul Preparatory Seoul, Seoul (Korea, Republic of)

2016-10-15

This paper introduces an integrated signal preconditioning and model prediction mainly by kernel functions. The performance and benefits of the methods are demonstrated by a case study with measurement data from a power plant and its components transient data. The developed methods will be applied as a part of monitoring massive or big data platform where human experts cannot detect the fault behaviors due to too large size of the measurements. Recent extensive efforts for on-line monitoring implementation insists that a big surprise in the modeling for predicting process variables was the extent of data quality problems in measurement data especially for data-driven modeling. Bad data for training will be learned as normal and can make significant degrade in prediction performance. For this reason, the quantity and quality of measurement data in modeling phase need special care. Bad quality data must be removed from training sets to the bad data considered as normal system behavior. This paper presented an integrated structure of supervisory system for monitoring the plants or sensors performance. The quality of the data-driven model is improved with a bilateral kernel filter for preprocessing of the noisy data. The prediction module is also based on kernel regression having the same basis with noise filter. The model structure is optimized by a grouping process with nonlinear Hoeffding correlation function.

Protein homology model refinement by large-scale energy optimization.

Science.gov (United States)

Park, Hahnbeom; Ovchinnikov, Sergey; Kim, David E; DiMaio, Frank; Baker, David

2018-03-20

Proteins fold to their lowest free-energy structures, and hence the most straightforward way to increase the accuracy of a partially incorrect protein structure model is to search for the lowest-energy nearby structure. This direct approach has met with little success for two reasons: first, energy function inaccuracies can lead to false energy minima, resulting in model degradation rather than improvement; and second, even with an accurate energy function, the search problem is formidable because the energy only drops considerably in the immediate vicinity of the global minimum, and there are a very large number of degrees of freedom. Here we describe a large-scale energy optimization-based refinement method that incorporates advances in both search and energy function accuracy that can substantially improve the accuracy of low-resolution homology models. The method refined low-resolution homology models into correct folds for 50 of 84 diverse protein families and generated improved models in recent blind structure prediction experiments. Analyses of the basis for these improvements reveal contributions from both the improvements in conformational sampling techniques and the energy function.

Comparing vector-based and Bayesian memory models using large-scale datasets: User-generated hashtag and tag prediction on Twitter and Stack Overflow.

Science.gov (United States)

Stanley, Clayton; Byrne, Michael D

2016-12-01

The growth of social media and user-created content on online sites provides unique opportunities to study models of human declarative memory. By framing the task of choosing a hashtag for a tweet and tagging a post on Stack Overflow as a declarative memory retrieval problem, 2 cognitively plausible declarative memory models were applied to millions of posts and tweets and evaluated on how accurately they predict a user's chosen tags. An ACT-R based Bayesian model and a random permutation vector-based model were tested on the large data sets. The results show that past user behavior of tag use is a strong predictor of future behavior. Furthermore, past behavior was successfully incorporated into the random permutation model that previously used only context. Also, ACT-R's attentional weight term was linked to an entropy-weighting natural language processing method used to attenuate high-frequency words (e.g., articles and prepositions). Word order was not found to be a strong predictor of tag use, and the random permutation model performed comparably to the Bayesian model without including word order. This shows that the strength of the random permutation model is not in the ability to represent word order, but rather in the way in which context information is successfully compressed. The results of the large-scale exploration show how the architecture of the 2 memory models can be modified to significantly improve accuracy, and may suggest task-independent general modifications that can help improve model fit to human data in a much wider range of domains. (PsycINFO Database Record (c) 2016 APA, all rights reserved).

Drivers and seasonal predictability of extreme wind speeds in the ECMWF System 4 and a statistical model

Science.gov (United States)

Walz, M. A.; Donat, M.; Leckebusch, G. C.

2017-12-01

As extreme wind speeds are responsible for large socio-economic losses in Europe, a skillful prediction would be of great benefit for disaster prevention as well as for the actuarial community. Here we evaluate patterns of large-scale atmospheric variability and the seasonal predictability of extreme wind speeds (e.g. >95th percentile) in the European domain in the dynamical seasonal forecast system ECMWF System 4, and compare to the predictability based on a statistical prediction model. The dominant patterns of atmospheric variability show distinct differences between reanalysis and ECMWF System 4, with most patterns in System 4 extended downstream in comparison to ERA-Interim. The dissimilar manifestations of the patterns within the two models lead to substantially different drivers associated with the occurrence of extreme winds in the respective model. While the ECMWF System 4 is shown to provide some predictive power over Scandinavia and the eastern Atlantic, only very few grid cells in the European domain have significant correlations for extreme wind speeds in System 4 compared to ERA-Interim. In contrast, a statistical model predicts extreme wind speeds during boreal winter in better agreement with the observations. Our results suggest that System 4 does not seem to capture the potential predictability of extreme winds that exists in the real world, and therefore fails to provide reliable seasonal predictions for lead months 2-4. This is likely related to the unrealistic representation of large-scale patterns of atmospheric variability. Hence our study points to potential improvements of dynamical prediction skill by improving the simulation of large-scale atmospheric dynamics.

Wind power prediction models

Science.gov (United States)

Levy, R.; Mcginness, H.

1976-01-01

Investigations were performed to predict the power available from the wind at the Goldstone, California, antenna site complex. The background for power prediction was derived from a statistical evaluation of available wind speed data records at this location and at nearby locations similarly situated within the Mojave desert. In addition to a model for power prediction over relatively long periods of time, an interim simulation model that produces sample wind speeds is described. The interim model furnishes uncorrelated sample speeds at hourly intervals that reproduce the statistical wind distribution at Goldstone. A stochastic simulation model to provide speed samples representative of both the statistical speed distributions and correlations is also discussed.

Inverse and Predictive Modeling

Energy Technology Data Exchange (ETDEWEB)

Syracuse, Ellen Marie [Los Alamos National Lab. (LANL), Los Alamos, NM (United States)

2017-09-27

The LANL Seismo-Acoustic team has a strong capability in developing data-driven models that accurately predict a variety of observations. These models range from the simple – one-dimensional models that are constrained by a single dataset and can be used for quick and efficient predictions – to the complex – multidimensional models that are constrained by several types of data and result in more accurate predictions. Team members typically build models of geophysical characteristics of Earth and source distributions at scales of 1 to 1000s of km, the techniques used are applicable for other types of physical characteristics at an even greater range of scales. The following cases provide a snapshot of some of the modeling work done by the Seismo- Acoustic team at LANL.

Archaeological predictive model set.

Science.gov (United States)

2015-03-01

This report is the documentation for Task 7 of the Statewide Archaeological Predictive Model Set. The goal of this project is to : develop a set of statewide predictive models to assist the planning of transportation projects. PennDOT is developing t...

Regression-based season-ahead drought prediction for southern Peru conditioned on large-scale climate variables

Science.gov (United States)

Mortensen, Eric; Wu, Shu; Notaro, Michael; Vavrus, Stephen; Montgomery, Rob; De Piérola, José; Sánchez, Carlos; Block, Paul

2018-01-01

Located at a complex topographic, climatic, and hydrologic crossroads, southern Peru is a semiarid region that exhibits high spatiotemporal variability in precipitation. The economic viability of the region hinges on this water, yet southern Peru is prone to water scarcity caused by seasonal meteorological drought. Meteorological droughts in this region are often triggered during El Niño episodes; however, other large-scale climate mechanisms also play a noteworthy role in controlling the region's hydrologic cycle. An extensive season-ahead precipitation prediction model is developed to help bolster the existing capacity of stakeholders to plan for and mitigate deleterious impacts of drought. In addition to existing climate indices, large-scale climatic variables, such as sea surface temperature, are investigated to identify potential drought predictors. A principal component regression framework is applied to 11 potential predictors to produce an ensemble forecast of regional January-March precipitation totals. Model hindcasts of 51 years, compared to climatology and another model conditioned solely on an El Niño-Southern Oscillation index, achieve notable skill and perform better for several metrics, including ranked probability skill score and a hit-miss statistic. The information provided by the developed model and ancillary modeling efforts, such as extending the lead time of and spatially disaggregating precipitation predictions to the local level as well as forecasting the number of wet-dry days per rainy season, may further assist regional stakeholders and policymakers in preparing for drought.

Web tools for predictive toxicology model building.

Science.gov (United States)

Jeliazkova, Nina

2012-07-01

The development and use of web tools in chemistry has accumulated more than 15 years of history already. Powered by the advances in the Internet technologies, the current generation of web systems are starting to expand into areas, traditional for desktop applications. The web platforms integrate data storage, cheminformatics and data analysis tools. The ease of use and the collaborative potential of the web is compelling, despite the challenges. The topic of this review is a set of recently published web tools that facilitate predictive toxicology model building. The focus is on software platforms, offering web access to chemical structure-based methods, although some of the frameworks could also provide bioinformatics or hybrid data analysis functionalities. A number of historical and current developments are cited. In order to provide comparable assessment, the following characteristics are considered: support for workflows, descriptor calculations, visualization, modeling algorithms, data management and data sharing capabilities, availability of GUI or programmatic access and implementation details. The success of the Web is largely due to its highly decentralized, yet sufficiently interoperable model for information access. The expected future convergence between cheminformatics and bioinformatics databases provides new challenges toward management and analysis of large data sets. The web tools in predictive toxicology will likely continue to evolve toward the right mix of flexibility, performance, scalability, interoperability, sets of unique features offered, friendly user interfaces, programmatic access for advanced users, platform independence, results reproducibility, curation and crowdsourcing utilities, collaborative sharing and secure access.

A Comparison Between Measured and Predicted Hydrodynamic Damping for a Jack-Up Rig Model

DEFF Research Database (Denmark)

Laursen, Thomas; Rohbock, Lars; Jensen, Jørgen Juncher

1996-01-01

An extensive set of measurements funded by the EU project Large Scale Facilities Program has been carried out on a model of a jack-up rig at the Danish Hydraulic Institute. The test serieswere conducted by MSC and include determination of base shears and overturning moments in both regular...... methods.In the comparison between the model test results and the theoretical predictions, thehydro-dynamic damping proves to be the most important uncertain parameter. It is shown thata relative large hydrodynamic damping must be assumed in the theoretical calculations in orderto predict the measured...

Outcome Prediction in Mathematical Models of Immune Response to Infection.

Directory of Open Access Journals (Sweden)

Manuel Mai

Full Text Available Clinicians need to predict patient outcomes with high accuracy as early as possible after disease inception. In this manuscript, we show that patient-to-patient variability sets a fundamental limit on outcome prediction accuracy for a general class of mathematical models for the immune response to infection. However, accuracy can be increased at the expense of delayed prognosis. We investigate several systems of ordinary differential equations (ODEs that model the host immune response to a pathogen load. Advantages of systems of ODEs for investigating the immune response to infection include the ability to collect data on large numbers of 'virtual patients', each with a given set of model parameters, and obtain many time points during the course of the infection. We implement patient-to-patient variability v in the ODE models by randomly selecting the model parameters from distributions with coefficients of variation v that are centered on physiological values. We use logistic regression with one-versus-all classification to predict the discrete steady-state outcomes of the system. We find that the prediction algorithm achieves near 100% accuracy for v = 0, and the accuracy decreases with increasing v for all ODE models studied. The fact that multiple steady-state outcomes can be obtained for a given initial condition, i.e. the basins of attraction overlap in the space of initial conditions, limits the prediction accuracy for v > 0. Increasing the elapsed time of the variables used to train and test the classifier, increases the prediction accuracy, while adding explicit external noise to the ODE models decreases the prediction accuracy. Our results quantify the competition between early prognosis and high prediction accuracy that is frequently encountered by clinicians.

The potential of large studies for building genetic risk prediction models

Science.gov (United States)

NCI scientists have developed a new paradigm to assess hereditary risk prediction in common diseases, such as prostate cancer. This genetic risk prediction concept is based on polygenic analysis—the study of a group of common DNA sequences, known as singl

Large tan β in gauge-mediated SUSY-breaking models

International Nuclear Information System (INIS)

Rattazzi, R.

1997-01-01

We explore some topics in the phenomenology of gauge-mediated SUSY-breaking scenarios having a large hierarchy of Higgs VEVs, v U /v D = tan β>>1. Some motivation for this scenario is first presented. We then use a systematic, analytic expansion (including some threshold corrections) to calculate the μ-parameter needed for proper electroweak breaking and the radiative corrections to the B-parameter, which fortuitously cancel at leading order. If B = 0 at the messenger scale then tan β is naturally large and calculable; we calculate it. We then confront this prediction with classical and quantum vacuum stability constraints arising from the Higgs-slepton potential, and indicate the preferred values of the top quark mass and messenger scale(s). The possibility of vacuum instability in a different direction yields an upper bound on the messenger mass scale complementary to the familiar bound from gravitino relic abundance. Next, we calculate the rate for b→sγ and show the possibility of large deviations (in the direction currently favored by experiment) from standard-model and small tan β predictions. Finally, we discuss the implications of these findings and their applicability to future, broader and more detailed investigations. (orig.)

Comparison of hard scattering models for particle production at large transverse momentum. 2

International Nuclear Information System (INIS)

Schiller, A.; Ilgenfritz, E.M.; Kripfganz, J.; Moehring, H.J.; Ranft, G.; Ranft, J.

1977-01-01

Single particle distributions of π + and π - at large transverse momentum are analysed using various hard collision models: qq → qq, qantiq → MantiM, qM → qM. The transverse momentum dependence at thetasub(cm) = 90 0 is well described in all models except qantiq → MantiM. This model has problems with the ratios (pp → π + +X)/(π +- p → π 0 +X). Presently available data on rapidity distributions of pions in π - p and pantip collisions are at rather low transverse momentum (however large xsub(perpendicular) = 2psub(perpendicular)/√s) where it is not obvious that hard collision models should dominate. The data, in particular the π - /π + asymmetry are well described by all models except qM → Mq (CIM). At large values of transverse momentum significant differences between the models are predicted. (author)

An improved large signal model of InP HEMTs

Science.gov (United States)

Li, Tianhao; Li, Wenjun; Liu, Jun

2018-05-01

An improved large signal model for InP HEMTs is proposed in this paper. The channel current and charge model equations are constructed based on the Angelov model equations. Both the equations for channel current and gate charge models were all continuous and high order drivable, and the proposed gate charge model satisfied the charge conservation. For the strong leakage induced barrier reduction effect of InP HEMTs, the Angelov current model equations are improved. The channel current model could fit DC performance of devices. A 2 × 25 μm × 70 nm InP HEMT device is used to demonstrate the extraction and validation of the model, in which the model has predicted the DC I–V, C–V and bias related S parameters accurately. Project supported by the National Natural Science Foundation of China (No. 61331006).

Subgrid-scale models for large-eddy simulation of rotating turbulent channel flows

Science.gov (United States)

Silvis, Maurits H.; Bae, Hyunji Jane; Trias, F. Xavier; Abkar, Mahdi; Moin, Parviz; Verstappen, Roel

2017-11-01

We aim to design subgrid-scale models for large-eddy simulation of rotating turbulent flows. Rotating turbulent flows form a challenging test case for large-eddy simulation due to the presence of the Coriolis force. The Coriolis force conserves the total kinetic energy while transporting it from small to large scales of motion, leading to the formation of large-scale anisotropic flow structures. The Coriolis force may also cause partial flow laminarization and the occurrence of turbulent bursts. Many subgrid-scale models for large-eddy simulation are, however, primarily designed to parametrize the dissipative nature of turbulent flows, ignoring the specific characteristics of transport processes. We, therefore, propose a new subgrid-scale model that, in addition to the usual dissipative eddy viscosity term, contains a nondissipative nonlinear model term designed to capture transport processes, such as those due to rotation. We show that the addition of this nonlinear model term leads to improved predictions of the energy spectra of rotating homogeneous isotropic turbulence as well as of the Reynolds stress anisotropy in spanwise-rotating plane-channel flows. This work is financed by the Netherlands Organisation for Scientific Research (NWO) under Project Number 613.001.212.

Confidence scores for prediction models

DEFF Research Database (Denmark)

Gerds, Thomas Alexander; van de Wiel, MA

2011-01-01

In medical statistics, many alternative strategies are available for building a prediction model based on training data. Prediction models are routinely compared by means of their prediction performance in independent validation data. If only one data set is available for training and validation,...

Particle production at large transverse momentum and hard collision models

International Nuclear Information System (INIS)

Ranft, G.; Ranft, J.

1977-04-01

The majority of the presently available experimental data is consistent with hard scattering models. Therefore the hard scattering model seems to be well established. There is good evidence for jets in large transverse momentum reactions as predicted by these models. The overall picture is however not yet well enough understood. We mention only the empirical hard scattering cross section introduced in most of the models, the lack of a deep theoretical understanding of the interplay between quark confinement and jet production, and the fact that we are not yet able to discriminate conclusively between the many proposed hard scattering models. The status of different hard collision models discussed in this paper is summarized. (author)

A dynamic globalization model for large eddy simulation of complex turbulent flow

Energy Technology Data Exchange (ETDEWEB)

Choi, Hae Cheon; Park, No Ma; Kim, Jin Seok [Seoul National Univ., Seoul (Korea, Republic of)

2005-07-01

A dynamic subgrid-scale model is proposed for large eddy simulation of turbulent flows in complex geometry. The eddy viscosity model by Vreman [Phys. Fluids, 16, 3670 (2004)] is considered as a base model. A priori tests with the original Vreman model show that it predicts the correct profile of subgrid-scale dissipation in turbulent channel flow but the optimal model coefficient is far from universal. Dynamic procedures of determining the model coefficient are proposed based on the 'global equilibrium' between the subgrid-scale dissipation and viscous dissipation. An important feature of the proposed procedures is that the model coefficient determined is globally constant in space but varies only in time. Large eddy simulations with the present dynamic model are conducted for forced isotropic turbulence, turbulent channel flow and flow over a sphere, showing excellent agreements with previous results.

Model Predictive Vibration Control Efficient Constrained MPC Vibration Control for Lightly Damped Mechanical Structures

CERN Document Server

Takács, Gergely

2012-01-01

Real-time model predictive controller (MPC) implementation in active vibration control (AVC) is often rendered difficult by fast sampling speeds and extensive actuator-deformation asymmetry. If the control of lightly damped mechanical structures is assumed, the region of attraction containing the set of allowable initial conditions requires a large prediction horizon, making the already computationally demanding on-line process even more complex. Model Predictive Vibration Control provides insight into the predictive control of lightly damped vibrating structures by exploring computationally efficient algorithms which are capable of low frequency vibration control with guaranteed stability and constraint feasibility. In addition to a theoretical primer on active vibration damping and model predictive control, Model Predictive Vibration Control provides a guide through the necessary steps in understanding the founding ideas of predictive control applied in AVC such as: ·         the implementation of ...

Zone modelling of the thermal performances of a large-scale bloom reheating furnace

International Nuclear Information System (INIS)

Tan, Chee-Keong; Jenkins, Joana; Ward, John; Broughton, Jonathan; Heeley, Andy

2013-01-01

This paper describes the development and comparison of a two- (2D) and three-dimensional (3D) mathematical models, based on the zone method of radiation analysis, to simulate the thermal performances of a large bloom reheating furnace. The modelling approach adopted in the current paper differs from previous work since it takes into account the net radiation interchanges between the top and bottom firing sections of the furnace and also allows for enthalpy exchange due to the flows of combustion products between these sections. The models were initially validated at two different furnace throughput rates using experimental and plant's model data supplied by Tata Steel. The results to-date demonstrated that the model predictions are in good agreement with measured heating profiles of the blooms encountered in the actual furnace. It was also found no significant differences between the predictions from the 2D and 3D models. Following the validation, the 2D model was then used to assess the impact of the furnace responses to changing throughput rate. It was found that the potential furnace response to changing throughput rate influences the settling time of the furnace to the next steady state operation. Overall the current work demonstrates the feasibility and practicality of zone modelling and its potential for incorporation into a model based furnace control system. - Highlights: ► 2D and 3D zone models of large-scale bloom reheating furnace. ► The models were validated with experimental and plant model data. ► Examine the transient furnace response to changing the furnace throughput rates. ► No significant differences found between the predictions from the 2D and 3D models.

A Personalized Predictive Framework for Multivariate Clinical Time Series via Adaptive Model Selection.

Science.gov (United States)

Liu, Zitao; Hauskrecht, Milos

2017-11-01

Building of an accurate predictive model of clinical time series for a patient is critical for understanding of the patient condition, its dynamics, and optimal patient management. Unfortunately, this process is not straightforward. First, patient-specific variations are typically large and population-based models derived or learned from many different patients are often unable to support accurate predictions for each individual patient. Moreover, time series observed for one patient at any point in time may be too short and insufficient to learn a high-quality patient-specific model just from the patient's own data. To address these problems we propose, develop and experiment with a new adaptive forecasting framework for building multivariate clinical time series models for a patient and for supporting patient-specific predictions. The framework relies on the adaptive model switching approach that at any point in time selects the most promising time series model out of the pool of many possible models, and consequently, combines advantages of the population, patient-specific and short-term individualized predictive models. We demonstrate that the adaptive model switching framework is very promising approach to support personalized time series prediction, and that it is able to outperform predictions based on pure population and patient-specific models, as well as, other patient-specific model adaptation strategies.

Breast cancer risks and risk prediction models.

Science.gov (United States)

Engel, Christoph; Fischer, Christine

2015-02-01

BRCA1/2 mutation carriers have a considerably increased risk to develop breast and ovarian cancer. The personalized clinical management of carriers and other at-risk individuals depends on precise knowledge of the cancer risks. In this report, we give an overview of the present literature on empirical cancer risks, and we describe risk prediction models that are currently used for individual risk assessment in clinical practice. Cancer risks show large variability between studies. Breast cancer risks are at 40-87% for BRCA1 mutation carriers and 18-88% for BRCA2 mutation carriers. For ovarian cancer, the risk estimates are in the range of 22-65% for BRCA1 and 10-35% for BRCA2. The contralateral breast cancer risk is high (10-year risk after first cancer 27% for BRCA1 and 19% for BRCA2). Risk prediction models have been proposed to provide more individualized risk prediction, using additional knowledge on family history, mode of inheritance of major genes, and other genetic and non-genetic risk factors. User-friendly software tools have been developed that serve as basis for decision-making in family counseling units. In conclusion, further assessment of cancer risks and model validation is needed, ideally based on prospective cohort studies. To obtain such data, clinical management of carriers and other at-risk individuals should always be accompanied by standardized scientific documentation.

Hydrogen combustion modelling in large-scale geometries

International Nuclear Information System (INIS)

Studer, E.; Beccantini, A.; Kudriakov, S.; Velikorodny, A.

2014-01-01

Hydrogen risk mitigation issues based on catalytic recombiners cannot exclude flammable clouds to be formed during the course of a severe accident in a Nuclear Power Plant. Consequences of combustion processes have to be assessed based on existing knowledge and state of the art in CFD combustion modelling. The Fukushima accidents have also revealed the need for taking into account the hydrogen explosion phenomena in risk management. Thus combustion modelling in a large-scale geometry is one of the remaining severe accident safety issues. At present day there doesn't exist a combustion model which can accurately describe a combustion process inside a geometrical configuration typical of the Nuclear Power Plant (NPP) environment. Therefore the major attention in model development has to be paid on the adoption of existing approaches or creation of the new ones capable of reliably predicting the possibility of the flame acceleration in the geometries of that type. A set of experiments performed previously in RUT facility and Heiss Dampf Reactor (HDR) facility is used as a validation database for development of three-dimensional gas dynamic model for the simulation of hydrogen-air-steam combustion in large-scale geometries. The combustion regimes include slow deflagration, fast deflagration, and detonation. Modelling is based on Reactive Discrete Equation Method (RDEM) where flame is represented as an interface separating reactants and combustion products. The transport of the progress variable is governed by different flame surface wrinkling factors. The results of numerical simulation are presented together with the comparisons, critical discussions and conclusions. (authors)

Modelling Morphological Response of Large Tidal Inlet Systems to Sea Level Rise

NARCIS (Netherlands)

Dissanayake, P.K.

2011-01-01

This dissertation qualitatively investigates the morphodynamic response of a large inlet system to IPCC projected relative sea level rise (RSLR). Adopted numerical approach (Delft3D) used a highly schematised model domain analogous to the Ameland inlet in the Dutch Wadden Sea. Predicted inlet

Thermal Storage Power Balancing with Model Predictive Control

DEFF Research Database (Denmark)

Halvgaard, Rasmus; Poulsen, Niels Kjølstad; Madsen, Henrik

2013-01-01

The method described in this paper balances power production and consumption with a large number of thermal loads. Linear controllers are used for the loads to track a temperature set point, while Model Predictive Control (MPC) and model estimation of the load behavior are used for coordination....... The total power consumption of all loads is controlled indirectly through a real-time price. The MPC incorporates forecasts of the power production and disturbances that influence the loads, e.g. time-varying weather forecasts, in order to react ahead of time. A simulation scenario demonstrates...

Model for the prediction of subsurface strata movement due to underground mining

Science.gov (United States)

Cheng, Jianwei; Liu, Fangyuan; Li, Siyuan

2017-12-01

The problem of ground control stability due to large underground mining operations is often associated with large movements and deformations of strata. It is a complicated problem, and can induce severe safety or environmental hazards either at the surface or in strata. Hence, knowing the subsurface strata movement characteristics, and making any subsidence predictions in advance, are desirable for mining engineers to estimate any damage likely to affect the ground surface or subsurface strata. Based on previous research findings, this paper broadly applies a surface subsidence prediction model based on the influence function method to subsurface strata, in order to predict subsurface stratum movement. A step-wise prediction model is proposed, to investigate the movement of underground strata. The model involves a dynamic iteration calculation process to derive the movements and deformations for each stratum layer; modifications to the influence method function are also made for more precise calculations. The critical subsidence parameters, incorporating stratum mechanical properties and the spatial relationship of interest at the mining level, are thoroughly considered, with the purpose of improving the reliability of input parameters. Such research efforts can be very helpful to mining engineers’ understanding of the moving behavior of all strata over underground excavations, and assist in making any damage mitigation plan. In order to check the reliability of the model, two methods are carried out and cross-validation applied. One is to use a borehole TV monitor recording to identify the progress of subsurface stratum bedding and caving in a coal mine, the other is to conduct physical modelling of the subsidence in underground strata. The results of these two methods are used to compare with theoretical results calculated by the proposed mathematical model. The testing results agree well with each other, and the acceptable accuracy and reliability of the

Gene prediction in metagenomic fragments: A large scale machine learning approach

Directory of Open Access Journals (Sweden)

Morgenstern Burkhard

2008-04-01

Full Text Available Abstract Background Metagenomics is an approach to the characterization of microbial genomes via the direct isolation of genomic sequences from the environment without prior cultivation. The amount of metagenomic sequence data is growing fast while computational methods for metagenome analysis are still in their infancy. In contrast to genomic sequences of single species, which can usually be assembled and analyzed by many available methods, a large proportion of metagenome data remains as unassembled anonymous sequencing reads. One of the aims of all metagenomic sequencing projects is the identification of novel genes. Short length, for example, Sanger sequencing yields on average 700 bp fragments, and unknown phylogenetic origin of most fragments require approaches to gene prediction that are different from the currently available methods for genomes of single species. In particular, the large size of metagenomic samples requires fast and accurate methods with small numbers of false positive predictions. Results We introduce a novel gene prediction algorithm for metagenomic fragments based on a two-stage machine learning approach. In the first stage, we use linear discriminants for monocodon usage, dicodon usage and translation initiation sites to extract features from DNA sequences. In the second stage, an artificial neural network combines these features with open reading frame length and fragment GC-content to compute the probability that this open reading frame encodes a protein. This probability is used for the classification and scoring of gene candidates. With large scale training, our method provides fast single fragment predictions with good sensitivity and specificity on artificially fragmented genomic DNA. Additionally, this method is able to predict translation initiation sites accurately and distinguishes complete from incomplete genes with high reliability. Conclusion Large scale machine learning methods are well-suited for gene

Multi-model comparison highlights consistency in predicted effect of warming on a semi-arid shrub

Science.gov (United States)

Renwick, Katherine M.; Curtis, Caroline; Kleinhesselink, Andrew R.; Schlaepfer, Daniel R.; Bradley, Bethany A.; Aldridge, Cameron L.; Poulter, Benjamin; Adler, Peter B.

2018-01-01

A number of modeling approaches have been developed to predict the impacts of climate change on species distributions, performance, and abundance. The stronger the agreement from models that represent different processes and are based on distinct and independent sources of information, the greater the confidence we can have in their predictions. Evaluating the level of confidence is particularly important when predictions are used to guide conservation or restoration decisions. We used a multi-model approach to predict climate change impacts on big sagebrush (Artemisia tridentata), the dominant plant species on roughly 43 million hectares in the western United States and a key resource for many endemic wildlife species. To evaluate the climate sensitivity of A. tridentata, we developed four predictive models, two based on empirically derived spatial and temporal relationships, and two that applied mechanistic approaches to simulate sagebrush recruitment and growth. This approach enabled us to produce an aggregate index of climate change vulnerability and uncertainty based on the level of agreement between models. Despite large differences in model structure, predictions of sagebrush response to climate change were largely consistent. Performance, as measured by change in cover, growth, or recruitment, was predicted to decrease at the warmest sites, but increase throughout the cooler portions of sagebrush's range. A sensitivity analysis indicated that sagebrush performance responds more strongly to changes in temperature than precipitation. Most of the uncertainty in model predictions reflected variation among the ecological models, raising questions about the reliability of forecasts based on a single modeling approach. Our results highlight the value of a multi-model approach in forecasting climate change impacts and uncertainties and should help land managers to maximize the value of conservation investments.

Predictability in the Epidemic-Type Aftershock Sequence model of interacting triggered seismicity

Science.gov (United States)

Helmstetter, AgnèS.; Sornette, Didier

2003-10-01

As part of an effort to develop a systematic methodology for earthquake forecasting, we use a simple model of seismicity on the basis of interacting events which may trigger a cascade of earthquakes, known as the Epidemic-Type Aftershock Sequence model (ETAS). The ETAS model is constructed on a bare (unrenormalized) Omori law, the Gutenberg-Richter law, and the idea that large events trigger more numerous aftershocks. For simplicity, we do not use the information on the spatial location of earthquakes and work only in the time domain. We demonstrate the essential role played by the cascade of triggered seismicity in controlling the rate of aftershock decay as well as the overall level of seismicity in the presence of a constant external seismicity source. We offer an analytical approach to account for the yet unobserved triggered seismicity adapted to the problem of forecasting future seismic rates at varying horizons from the present. Tests presented on synthetic catalogs validate strongly the importance of taking into account all the cascades of still unobserved triggered events in order to predict correctly the future level of seismicity beyond a few minutes. We find a strong predictability if one accepts to predict only a small fraction of the large-magnitude targets. Specifically, we find a prediction gain (defined as the ratio of the fraction of predicted events over the fraction of time in alarms) equal to 21 for a fraction of alarm of 1%, a target magnitude M ≥ 6, an update time of 0.5 days between two predictions, and for realistic parameters of the ETAS model. However, the probability gains degrade fast when one attempts to predict a larger fraction of the targets. This is because a significant fraction of events remain uncorrelated from past seismicity. This delineates the fundamental limits underlying forecasting skills, stemming from an intrinsic stochastic component in these interacting triggered seismicity models. Quantitatively, the fundamental

The large-scale peculiar velocity field in flat models of the universe

International Nuclear Information System (INIS)

Vittorio, N.; Turner, M.S.

1986-10-01

The inflationary Universe scenario predicts a flat Universe and both adiabatic and isocurvature primordial density perturbations with the Zel'dovich spectrum. The two simplest realizations, models dominated by hot or cold dark matter, seem to be in conflict with observations. Flat models are examined with two components of mass density, where one of the components of mass density is smoothly distributed and the large-scale (≥10h -1 MpC) peculiar velocity field for these models is considered. For the smooth component relativistic particles, a relic cosmological term, and light strings are considered. At present the observational situation is unsettled; but, in principle, the large-scale peculiar velocity field is very powerful discriminator between these different models. 61 refs

Predictability of the recent slowdown and subsequent recovery of large-scale surface warming using statistical methods

Science.gov (United States)

Mann, Michael E.; Steinman, Byron A.; Miller, Sonya K.; Frankcombe, Leela M.; England, Matthew H.; Cheung, Anson H.

2016-04-01

The temporary slowdown in large-scale surface warming during the early 2000s has been attributed to both external and internal sources of climate variability. Using semiempirical estimates of the internal low-frequency variability component in Northern Hemisphere, Atlantic, and Pacific surface temperatures in concert with statistical hindcast experiments, we investigate whether the slowdown and its recent recovery were predictable. We conclude that the internal variability of the North Pacific, which played a critical role in the slowdown, does not appear to have been predictable using statistical forecast methods. An additional minor contribution from the North Atlantic, by contrast, appears to exhibit some predictability. While our analyses focus on combining semiempirical estimates of internal climatic variability with statistical hindcast experiments, possible implications for initialized model predictions are also discussed.

A new risk prediction model for critical care: the Intensive Care National Audit & Research Centre (ICNARC) model.

Science.gov (United States)

Harrison, David A; Parry, Gareth J; Carpenter, James R; Short, Alasdair; Rowan, Kathy

2007-04-01

To develop a new model to improve risk prediction for admissions to adult critical care units in the UK. Prospective cohort study. The setting was 163 adult, general critical care units in England, Wales, and Northern Ireland, December 1995 to August 2003. Patients were 216,626 critical care admissions. None. The performance of different approaches to modeling physiologic measurements was evaluated, and the best methods were selected to produce a new physiology score. This physiology score was combined with other information relating to the critical care admission-age, diagnostic category, source of admission, and cardiopulmonary resuscitation before admission-to develop a risk prediction model. Modeling interactions between diagnostic category and physiology score enabled the inclusion of groups of admissions that are frequently excluded from risk prediction models. The new model showed good discrimination (mean c index 0.870) and fit (mean Shapiro's R 0.665, mean Brier's score 0.132) in 200 repeated validation samples and performed well when compared with recalibrated versions of existing published risk prediction models in the cohort of patients eligible for all models. The hypothesis of perfect fit was rejected for all models, including the Intensive Care National Audit & Research Centre (ICNARC) model, as is to be expected in such a large cohort. The ICNARC model demonstrated better discrimination and overall fit than existing risk prediction models, even following recalibration of these models. We recommend it be used to replace previously published models for risk adjustment in the UK.

Model predictive control of a wind turbine modelled in Simpack

International Nuclear Information System (INIS)

Jassmann, U; Matzke, D; Reiter, M; Abel, D; Berroth, J; Schelenz, R; Jacobs, G

2014-01-01

Wind turbines (WT) are steadily growing in size to increase their power production, which also causes increasing loads acting on the turbine's components. At the same time large structures, such as the blades and the tower get more flexible. To minimize this impact, the classical control loops for keeping the power production in an optimum state are more and more extended by load alleviation strategies. These additional control loops can be unified by a multiple-input multiple-output (MIMO) controller to achieve better balancing of tuning parameters. An example for MIMO control, which has been paid more attention to recently by wind industry, is Model Predictive Control (MPC). In a MPC framework a simplified model of the WT is used to predict its controlled outputs. Based on a user-defined cost function an online optimization calculates the optimal control sequence. Thereby MPC can intrinsically incorporate constraints e.g. of actuators. Turbine models used for calculation within the MPC are typically simplified. For testing and verification usually multi body simulations, such as FAST, BLADED or FLEX5 are used to model system dynamics, but they are still limited in the number of degrees of freedom (DOF). Detailed information about load distribution (e.g. inside the gearbox) cannot be provided by such models. In this paper a Model Predictive Controller is presented and tested in a co-simulation with SlMPACK, a multi body system (MBS) simulation framework used for detailed load analysis. The analysis are performed on the basis of the IME6.0 MBS WT model, described in this paper. It is based on the rotor of the NREL 5MW WT and consists of a detailed representation of the drive train. This takes into account a flexible main shaft and its main bearings with a planetary gearbox, where all components are modelled flexible, as well as a supporting flexible main frame. The wind loads are simulated using the NREL AERODYN v13 code which has been implemented as a routine

Model predictive control of a wind turbine modelled in Simpack

Science.gov (United States)

Jassmann, U.; Berroth, J.; Matzke, D.; Schelenz, R.; Reiter, M.; Jacobs, G.; Abel, D.

2014-06-01

Wind turbines (WT) are steadily growing in size to increase their power production, which also causes increasing loads acting on the turbine's components. At the same time large structures, such as the blades and the tower get more flexible. To minimize this impact, the classical control loops for keeping the power production in an optimum state are more and more extended by load alleviation strategies. These additional control loops can be unified by a multiple-input multiple-output (MIMO) controller to achieve better balancing of tuning parameters. An example for MIMO control, which has been paid more attention to recently by wind industry, is Model Predictive Control (MPC). In a MPC framework a simplified model of the WT is used to predict its controlled outputs. Based on a user-defined cost function an online optimization calculates the optimal control sequence. Thereby MPC can intrinsically incorporate constraints e.g. of actuators. Turbine models used for calculation within the MPC are typically simplified. For testing and verification usually multi body simulations, such as FAST, BLADED or FLEX5 are used to model system dynamics, but they are still limited in the number of degrees of freedom (DOF). Detailed information about load distribution (e.g. inside the gearbox) cannot be provided by such models. In this paper a Model Predictive Controller is presented and tested in a co-simulation with SlMPACK, a multi body system (MBS) simulation framework used for detailed load analysis. The analysis are performed on the basis of the IME6.0 MBS WT model, described in this paper. It is based on the rotor of the NREL 5MW WT and consists of a detailed representation of the drive train. This takes into account a flexible main shaft and its main bearings with a planetary gearbox, where all components are modelled flexible, as well as a supporting flexible main frame. The wind loads are simulated using the NREL AERODYN v13 code which has been implemented as a routine to

Methods for Prediction of Steel Temperature Curve in the Whole Process of a Localized Fire in Large Spaces

Directory of Open Access Journals (Sweden)

Zhang Guowei

2014-01-01

Full Text Available Based on a full-scale bookcase fire experiment, a fire development model is proposed for the whole process of localized fires in large-space buildings. We found that for localized fires in large-space buildings full of wooden combustible materials the fire growing phases can be simplified into a t2 fire with a 0.0346 kW/s2 fire growth coefficient. FDS technology is applied to study the smoke temperature curve for a 2 MW to 25 MW fire occurring within a large space with a height of 6 m to 12 m and a building area of 1 500 m2 to 10 000 m2 based on the proposed fire development model. Through the analysis of smoke temperature in various fire scenarios, a new approach is proposed to predict the smoke temperature curve. Meanwhile, a modified model of steel temperature development in localized fire is built. In the modified model, the localized fire source is treated as a point fire source to evaluate the flame net heat flux to steel. The steel temperature curve in the whole process of a localized fire could be accurately predicted by the above findings. These conclusions obtained in this paper could provide valuable reference to fire simulation, hazard assessment, and fire protection design.

Predictive Models for Carcinogenicity and Mutagenicity ...

Science.gov (United States)

Mutagenicity and carcinogenicity are endpoints of major environmental and regulatory concern. These endpoints are also important targets for development of alternative methods for screening and prediction due to the large number of chemicals of potential concern and the tremendous cost (in time, money, animals) of rodent carcinogenicity bioassays. Both mutagenicity and carcinogenicity involve complex, cellular processes that are only partially understood. Advances in technologies and generation of new data will permit a much deeper understanding. In silico methods for predicting mutagenicity and rodent carcinogenicity based on chemical structural features, along with current mutagenicity and carcinogenicity data sets, have performed well for local prediction (i.e., within specific chemical classes), but are less successful for global prediction (i.e., for a broad range of chemicals). The predictivity of in silico methods can be improved by improving the quality of the data base and endpoints used for modelling. In particular, in vitro assays for clastogenicity need to be improved to reduce false positives (relative to rodent carcinogenicity) and to detect compounds that do not interact directly with DNA or have epigenetic activities. New assays emerging to complement or replace some of the standard assays include VitotoxTM, GreenScreenGC, and RadarScreen. The needs of industry and regulators to assess thousands of compounds necessitate the development of high-t

Wind Speed Prediction Using a Univariate ARIMA Model and a Multivariate NARX Model

Directory of Open Access Journals (Sweden)

Erasmo Cadenas

2016-02-01

Full Text Available Two on step ahead wind speed forecasting models were compared. A univariate model was developed using a linear autoregressive integrated moving average (ARIMA. This method’s performance is well studied for a large number of prediction problems. The other is a multivariate model developed using a nonlinear autoregressive exogenous artificial neural network (NARX. This uses the variables: barometric pressure, air temperature, wind direction and solar radiation or relative humidity, as well as delayed wind speed. Both models were developed from two databases from two sites: an hourly average measurements database from La Mata, Oaxaca, Mexico, and a ten minute average measurements database from Metepec, Hidalgo, Mexico. The main objective was to compare the impact of the various meteorological variables on the performance of the multivariate model of wind speed prediction with respect to the high performance univariate linear model. The NARX model gave better results with improvements on the ARIMA model of between 5.5% and 10. 6% for the hourly database and of between 2.3% and 12.8% for the ten minute database for mean absolute error and mean squared error, respectively.

Hybrid Reynolds-Averaged/Large Eddy Simulation of the Flow in a Model SCRamjet Cavity Flameholder

Science.gov (United States)

Baurle, R. A.

2016-01-01

Steady-state and scale-resolving simulations have been performed for flow in and around a model scramjet combustor flameholder. Experimental data available for this configuration include velocity statistics obtained from particle image velocimetry. Several turbulence models were used for the steady-state Reynolds-averaged simulations which included both linear and non-linear eddy viscosity models. The scale-resolving simulations used a hybrid Reynolds-averaged/large eddy simulation strategy that is designed to be a large eddy simulation everywhere except in the inner portion (log layer and below) of the boundary layer. Hence, this formulation can be regarded as a wall-modeled large eddy simulation. This e ort was undertaken to not only assess the performance of the hybrid Reynolds-averaged / large eddy simulation modeling approach in a flowfield of interest to the scramjet research community, but to also begin to understand how this capability can best be used to augment standard Reynolds-averaged simulations. The numerical errors were quantified for the steady-state simulations, and at least qualitatively assessed for the scale-resolving simulations prior to making any claims of predictive accuracy relative to the measurements. The steady-state Reynolds-averaged results displayed a high degree of variability when comparing the flameholder fuel distributions obtained from each turbulence model. This prompted the consideration of applying the higher-fidelity scale-resolving simulations as a surrogate "truth" model to calibrate the Reynolds-averaged closures in a non-reacting setting prior to their use for the combusting simulations. In general, the Reynolds-averaged velocity profile predictions at the lowest fueling level matched the particle imaging measurements almost as well as was observed for the non-reacting condition. However, the velocity field predictions proved to be more sensitive to the flameholder fueling rate than was indicated in the measurements.

Accurate and dynamic predictive model for better prediction in medicine and healthcare.

Science.gov (United States)

Alanazi, H O; Abdullah, A H; Qureshi, K N; Ismail, A S

2018-05-01

Information and communication technologies (ICTs) have changed the trend into new integrated operations and methods in all fields of life. The health sector has also adopted new technologies to improve the systems and provide better services to customers. Predictive models in health care are also influenced from new technologies to predict the different disease outcomes. However, still, existing predictive models have suffered from some limitations in terms of predictive outcomes performance. In order to improve predictive model performance, this paper proposed a predictive model by classifying the disease predictions into different categories. To achieve this model performance, this paper uses traumatic brain injury (TBI) datasets. TBI is one of the serious diseases worldwide and needs more attention due to its seriousness and serious impacts on human life. The proposed predictive model improves the predictive performance of TBI. The TBI data set is developed and approved by neurologists to set its features. The experiment results show that the proposed model has achieved significant results including accuracy, sensitivity, and specificity.

Cloud Based Metalearning System for Predictive Modeling of Biomedical Data

Directory of Open Access Journals (Sweden)

Milan Vukićević

2014-01-01

Full Text Available Rapid growth and storage of biomedical data enabled many opportunities for predictive modeling and improvement of healthcare processes. On the other side analysis of such large amounts of data is a difficult and computationally intensive task for most existing data mining algorithms. This problem is addressed by proposing a cloud based system that integrates metalearning framework for ranking and selection of best predictive algorithms for data at hand and open source big data technologies for analysis of biomedical data.

The BREAST-V: a unifying predictive formula for volume assessment in small, medium, and large breasts.

Science.gov (United States)

Longo, Benedetto; Farcomeni, Alessio; Ferri, Germano; Campanale, Antonella; Sorotos, Micheal; Santanelli, Fabio

2013-07-01

Breast volume assessment enhances preoperative planning of both aesthetic and reconstructive procedures, helping the surgeon in the decision-making process of shaping the breast. Numerous methods of breast size determination are currently reported but are limited by methodologic flaws and variable estimations. The authors aimed to develop a unifying predictive formula for volume assessment in small to large breasts based on anthropomorphic values. Ten anthropomorphic breast measurements and direct volumes of 108 mastectomy specimens from 88 women were collected prospectively. The authors performed a multivariate regression to build the optimal model for development of the predictive formula. The final model was then internally validated. A previously published formula was used as a reference. Mean (±SD) breast weight was 527.9 ± 227.6 g (range, 150 to 1250 g). After model selection, sternal notch-to-nipple, inframammary fold-to-nipple, and inframammary fold-to-fold projection distances emerged as the most important predictors. The resulting formula (the BREAST-V) showed an adjusted R of 0.73. The estimated expected absolute error on new breasts is 89.7 g (95 percent CI, 62.4 to 119.1 g) and the expected relative error is 18.4 percent (95 percent CI, 12.9 to 24.3 percent). Application of reference formula on the sample yielded worse predictions than those derived by the formula, showing an R of 0.55. The BREAST-V is a reliable tool for predicting small to large breast volumes accurately for use as a complementary device in surgeon evaluation. An app entitled BREAST-V for both iOS and Android devices is currently available for free download in the Apple App Store and Google Play Store. Diagnostic, II.

ARMA modelling of neutron stochastic processes with large measurement noise

International Nuclear Information System (INIS)

Zavaljevski, N.; Kostic, Lj.; Pesic, M.

1994-01-01

An autoregressive moving average (ARMA) model of the neutron fluctuations with large measurement noise is derived from langevin stochastic equations and validated using time series data obtained during prompt neutron decay constant measurements at the zero power reactor RB in Vinca. Model parameters are estimated using the maximum likelihood (ML) off-line algorithm and an adaptive pole estimation algorithm based on the recursive prediction error method (RPE). The results show that subcriticality can be determined from real data with high measurement noise using much shorter statistical sample than in standard methods. (author)

Modeling of large-scale oxy-fuel combustion processes

DEFF Research Database (Denmark)

Yin, Chungen

2012-01-01

Quite some studies have been conducted in order to implement oxy-fuel combustion with flue gas recycle in conventional utility boilers as an effective effort of carbon capture and storage. However, combustion under oxy-fuel conditions is significantly different from conventional air-fuel firing......, among which radiative heat transfer under oxy-fuel conditions is one of the fundamental issues. This paper demonstrates the nongray-gas effects in modeling of large-scale oxy-fuel combustion processes. Oxy-fuel combustion of natural gas in a 609MW utility boiler is numerically studied, in which...... calculation of the oxy-fuel WSGGM remarkably over-predicts the radiative heat transfer to the furnace walls and under-predicts the gas temperature at the furnace exit plane, which also result in a higher incomplete combustion in the gray calculation. Moreover, the gray and non-gray calculations of the same...

Adapting SWAT hillslope erosion model to predict sediment concentrations and yields in large Basins.

Science.gov (United States)

Vigiak, Olga; Malagó, Anna; Bouraoui, Fayçal; Vanmaercke, Matthias; Poesen, Jean

2015-12-15

The Soil and Water Assessment Tool (SWAT) is used worldwide for water quality assessment and planning. This paper aimed to assess and adapt SWAT hillslope sediment yield model (Modified Universal Soil Loss Equation, MUSLE) for applications in large basins, i.e. when spatial data is coarse and model units are large; and to develop a robust sediment calibration method for large regions. The Upper Danube Basin (132,000km(2)) was used as case study representative of large European Basins. The MUSLE was modified to reduce sensitivity of sediment yields to the Hydrologic Response Unit (HRU) size, and to identify appropriate algorithms for estimating hillslope length (L) and slope-length factor (LS). HRUs gross erosion was broadly calibrated against plot data and soil erosion map estimates. Next, mean annual SWAT suspended sediment concentrations (SSC, mg/L) were calibrated and validated against SSC data at 55 gauging stations (622 station-years). SWAT annual specific sediment yields in subbasin reaches (RSSY, t/km(2)/year) were compared to yields measured at 33 gauging stations (87station-years). The best SWAT configuration combined a MUSLE equation modified by the introduction of a threshold area of 0.01km(2) where L and LS were estimated with flow accumulation algorithms. For this configuration, the SSC residual interquartile was less than +/-15mg/L both for the calibration (1995-2004) and the validation (2005-2009) periods. The mean SSC percent bias for 1995-2009 was 24%. RSSY residual interquartile was within +/-10t/km(2)/year, with a mean RSSY percent bias of 12%. Residuals showed no bias with respect to drainage area, slope, or spatial distribution. The use of multiple data types at multiple sites enabled robust simulation of sediment concentrations and yields of the region. The MUSLE modifications are recommended for use in large basins. Based on SWAT simulations, we present a sediment budget for the Upper Danube Basin. Copyright © 2015. Published by Elsevier B.V.

Multiresolution comparison of precipitation datasets for large-scale models

Science.gov (United States)

Chun, K. P.; Sapriza Azuri, G.; Davison, B.; DeBeer, C. M.; Wheater, H. S.

2014-12-01

Gridded precipitation datasets are crucial for driving large-scale models which are related to weather forecast and climate research. However, the quality of precipitation products is usually validated individually. Comparisons between gridded precipitation products along with ground observations provide another avenue for investigating how the precipitation uncertainty would affect the performance of large-scale models. In this study, using data from a set of precipitation gauges over British Columbia and Alberta, we evaluate several widely used North America gridded products including the Canadian Gridded Precipitation Anomalies (CANGRD), the National Center for Environmental Prediction (NCEP) reanalysis, the Water and Global Change (WATCH) project, the thin plate spline smoothing algorithms (ANUSPLIN) and Canadian Precipitation Analysis (CaPA). Based on verification criteria for various temporal and spatial scales, results provide an assessment of possible applications for various precipitation datasets. For long-term climate variation studies (~100 years), CANGRD, NCEP, WATCH and ANUSPLIN have different comparative advantages in terms of their resolution and accuracy. For synoptic and mesoscale precipitation patterns, CaPA provides appealing performance of spatial coherence. In addition to the products comparison, various downscaling methods are also surveyed to explore new verification and bias-reduction methods for improving gridded precipitation outputs for large-scale models.

Dynamic Loads and Wake Prediction for Large Wind Turbines Based on Free Wake Method

Institute of Scientific and Technical Information of China (English)

Cao Jiufa; Wang Tongguang; Long Hui; Ke Shitang; Xu Bofeng

2015-01-01

With large scale wind turbines ,the issue of aerodynamic elastic response is even more significant on dy-namic behaviour of the system .Unsteady free vortex wake method is proposed to calculate the shape of wake and aerodynamic load .Considering the effect of aerodynamic load ,inertial load and gravity load ,the decoupling dy-namic equations are established by using finite element method in conjunction of the modal method and equations are solved numerically by Newmark approach .Finally ,the numerical simulation of a large scale wind turbine is performed through coupling the free vortex wake modelling with structural modelling .The results show that this coupling model can predict the flexible wind turbine dynamic characteristics effectively and efficiently .Under the influence of the gravitational force ,the dynamic response of flapwise direction contributes to the dynamic behavior of edgewise direction under the operational condition of steady wind speed .The difference in dynamic response be-tween the flexible and rigid wind turbines manifests when the aerodynamics/structure coupling effect is of signifi-cance in both wind turbine design and performance calculation .

An approach to model validation and model-based prediction -- polyurethane foam case study.

Energy Technology Data Exchange (ETDEWEB)

Dowding, Kevin J.; Rutherford, Brian Milne

2003-07-01

Enhanced software methodology and improved computing hardware have advanced the state of simulation technology to a point where large physics-based codes can be a major contributor in many systems analyses. This shift toward the use of computational methods has brought with it new research challenges in a number of areas including characterization of uncertainty, model validation, and the analysis of computer output. It is these challenges that have motivated the work described in this report. Approaches to and methods for model validation and (model-based) prediction have been developed recently in the engineering, mathematics and statistical literatures. In this report we have provided a fairly detailed account of one approach to model validation and prediction applied to an analysis investigating thermal decomposition of polyurethane foam. A model simulates the evolution of the foam in a high temperature environment as it transforms from a solid to a gas phase. The available modeling and experimental results serve as data for a case study focusing our model validation and prediction developmental efforts on this specific thermal application. We discuss several elements of the ''philosophy'' behind the validation and prediction approach: (1) We view the validation process as an activity applying to the use of a specific computational model for a specific application. We do acknowledge, however, that an important part of the overall development of a computational simulation initiative is the feedback provided to model developers and analysts associated with the application. (2) We utilize information obtained for the calibration of model parameters to estimate the parameters and quantify uncertainty in the estimates. We rely, however, on validation data (or data from similar analyses) to measure the variability that contributes to the uncertainty in predictions for specific systems or units (unit-to-unit variability). (3) We perform statistical

Comparing large-scale hydrological model predictions with observed streamflow in the Pacific Northwest: effects of climate and groundwater

Science.gov (United States)

Mohammad Safeeq; Guillaume S. Mauger; Gordon E. Grant; Ivan Arismendi; Alan F. Hamlet; Se-Yeun Lee

2014-01-01

Assessing uncertainties in hydrologic models can improve accuracy in predicting future streamflow. Here, simulated streamflows using the Variable Infiltration Capacity (VIC) model at coarse (1/16°) and fine (1/120°) spatial resolutions were evaluated against observed streamflows from 217 watersheds. In...

Prediction of welding residual distortions of large structures using a local/global approach

International Nuclear Information System (INIS)

Duan, Y. G.; Bergheau, J. M.; Vincent, Y.; Boitour, F.; Leblond, J. B.

2007-01-01

Prediction of welding residual distortions is more difficult than that of the microstructure and residual stresses. On the one hand, a fine mesh (often 3D) has to be used in the heat affected zone for the sake of the sharp variations of thermal, metallurgical and mechanical fields in this region. On the other hand, the whole structure is required to be meshed for the calculation of residual distortions. But for large structures, a 3D mesh is inconceivable caused by the costs of the calculation. Numerous methods have been developed to reduce the size of models. A local/global approach has been proposed to determine the welding residual distortions of large structures. The plastic strains and the microstructure due to welding are supposed can be determined from a local 3D model which concerns only the weld and its vicinity. They are projected as initial strains into a global 3D model which consists of the whole structure and obviously much less fine in the welded zone than the local model. The residual distortions are then calculated using a simple elastic analysis, which makes this method particularly effective in an industrial context. The aim of this article is to present the principle of the local/global approach then show the capacity of this method in an industrial context and finally study the definition of the local model

The Spike-and-Slab Lasso Generalized Linear Models for Prediction and Associated Genes Detection.

Science.gov (United States)

Tang, Zaixiang; Shen, Yueping; Zhang, Xinyan; Yi, Nengjun

2017-01-01

Large-scale "omics" data have been increasingly used as an important resource for prognostic prediction of diseases and detection of associated genes. However, there are considerable challenges in analyzing high-dimensional molecular data, including the large number of potential molecular predictors, limited number of samples, and small effect of each predictor. We propose new Bayesian hierarchical generalized linear models, called spike-and-slab lasso GLMs, for prognostic prediction and detection of associated genes using large-scale molecular data. The proposed model employs a spike-and-slab mixture double-exponential prior for coefficients that can induce weak shrinkage on large coefficients, and strong shrinkage on irrelevant coefficients. We have developed a fast and stable algorithm to fit large-scale hierarchal GLMs by incorporating expectation-maximization (EM) steps into the fast cyclic coordinate descent algorithm. The proposed approach integrates nice features of two popular methods, i.e., penalized lasso and Bayesian spike-and-slab variable selection. The performance of the proposed method is assessed via extensive simulation studies. The results show that the proposed approach can provide not only more accurate estimates of the parameters, but also better prediction. We demonstrate the proposed procedure on two cancer data sets: a well-known breast cancer data set consisting of 295 tumors, and expression data of 4919 genes; and the ovarian cancer data set from TCGA with 362 tumors, and expression data of 5336 genes. Our analyses show that the proposed procedure can generate powerful models for predicting outcomes and detecting associated genes. The methods have been implemented in a freely available R package BhGLM (http://www.ssg.uab.edu/bhglm/). Copyright © 2017 by the Genetics Society of America.

A maximum entropy model for predicting wild boar distribution in Spain

Directory of Open Access Journals (Sweden)

Jaime Bosch

2014-09-01

Full Text Available Wild boar (Sus scrofa populations in many areas of the Palearctic including the Iberian Peninsula have grown continuously over the last century. This increase has led to numerous different types of conflicts due to the damage these mammals can cause to agriculture, the problems they create in the conservation of natural areas, and the threat they pose to animal health. In the context of both wildlife management and the design of health programs for disease control, it is essential to know how wild boar are distributed on a large spatial scale. Given that the quantifying of the distribution of wild species using census techniques is virtually impossible in the case of large-scale studies, modeling techniques have thus to be used instead to estimate animals’ distributions, densities, and abundances. In this study, the potential distribution of wild boar in Spain was predicted by integrating data of presence and environmental variables into a MaxEnt approach. We built and tested models using 100 bootstrapped replicates. For each replicate or simulation, presence data was divided into two subsets that were used for model fitting (60% of the data and cross-validation (40% of the data. The final model was found to be accurate with an area under the receiver operating characteristic curve (AUC value of 0.79. Six explanatory variables for predicting wild boar distribution were identified on the basis of the percentage of their contribution to the model. The model exhibited a high degree of predictive accuracy, which has been confirmed by its agreement with satellite images and field surveys.

Parallel Computation of RCS of Electrically Large Platform with Coatings Modeled with NURBS Surfaces

Directory of Open Access Journals (Sweden)

Ying Yan

2012-01-01

Full Text Available The significance of Radar Cross Section (RCS in the military applications makes its prediction an important problem. This paper uses large-scale parallel Physical Optics (PO to realize the fast computation of RCS to electrically large targets, which are modeled by Non-Uniform Rational B-Spline (NURBS surfaces and coated with dielectric materials. Some numerical examples are presented to validate this paper’s method. In addition, 1024 CPUs are used in Shanghai Supercomputer Center (SSC to perform the simulation of a model with the maximum electrical size 1966.7 λ for the first time in China. From which, it can be found that this paper’s method can greatly speed the calculation and is capable of solving the real-life problem of RCS prediction.

Structural habitat predicts functional dispersal habitat of a large carnivore: how leopards change spots.

Science.gov (United States)

Fattebert, Julien; Robinson, Hugh S; Balme, Guy; Slotow, Rob; Hunter, Luke

2015-10-01

Natal dispersal promotes inter-population linkage, and is key to spatial distribution of populations. Degradation of suitable landscape structures beyond the specific threshold of an individual's ability to disperse can therefore lead to disruption of functional landscape connectivity and impact metapopulation function. Because it ignores behavioral responses of individuals, structural connectivity is easier to assess than functional connectivity and is often used as a surrogate for landscape connectivity modeling. However using structural resource selection models as surrogate for modeling functional connectivity through dispersal could be erroneous. We tested how well a second-order resource selection function (RSF) models (structural connectivity), based on GPS telemetry data from resident adult leopard (Panthera pardus L.), could predict subadult habitat use during dispersal (functional connectivity). We created eight non-exclusive subsets of the subadult data based on differing definitions of dispersal to assess the predictive ability of our adult-based RSF model extrapolated over a broader landscape. Dispersing leopards used habitats in accordance with adult selection patterns, regardless of the definition of dispersal considered. We demonstrate that, for a wide-ranging apex carnivore, functional connectivity through natal dispersal corresponds to structural connectivity as modeled by a second-order RSF. Mapping of the adult-based habitat classes provides direct visualization of the potential linkages between populations, without the need to model paths between a priori starting and destination points. The use of such landscape scale RSFs may provide insight into predicting suitable dispersal habitat peninsulas in human-dominated landscapes where mitigation of human-wildlife conflict should be focused. We recommend the use of second-order RSFs for landscape conservation planning and propose a similar approach to the conservation of other wide-ranging large

Chaotic advection at large Péclet number: Electromagnetically driven experiments, numerical simulations, and theoretical predictions

International Nuclear Information System (INIS)

Figueroa, Aldo; Meunier, Patrice; Villermaux, Emmanuel; Cuevas, Sergio; Ramos, Eduardo

2014-01-01

We present a combination of experiment, theory, and modelling on laminar mixing at large Péclet number. The flow is produced by oscillating electromagnetic forces in a thin electrolytic fluid layer, leading to oscillating dipoles, quadrupoles, octopoles, and disordered flows. The numerical simulations are based on the Diffusive Strip Method (DSM) which was recently introduced (P. Meunier and E. Villermaux, “The diffusive strip method for scalar mixing in two-dimensions,” J. Fluid Mech. 662, 134–172 (2010)) to solve the advection-diffusion problem by combining Lagrangian techniques and theoretical modelling of the diffusion. Numerical simulations obtained with the DSM are in reasonable agreement with quantitative dye visualization experiments of the scalar fields. A theoretical model based on log-normal Probability Density Functions (PDFs) of stretching factors, characteristic of homogeneous turbulence in the Batchelor regime, allows to predict the PDFs of scalar in agreement with numerical and experimental results. This model also indicates that the PDFs of scalar are asymptotically close to log-normal at late stages, except for the large concentration levels which correspond to low stretching factors

Fast integration-based prediction bands for ordinary differential equation models.

Science.gov (United States)

Hass, Helge; Kreutz, Clemens; Timmer, Jens; Kaschek, Daniel

2016-04-15

To gain a deeper understanding of biological processes and their relevance in disease, mathematical models are built upon experimental data. Uncertainty in the data leads to uncertainties of the model's parameters and in turn to uncertainties of predictions. Mechanistic dynamic models of biochemical networks are frequently based on nonlinear differential equation systems and feature a large number of parameters, sparse observations of the model components and lack of information in the available data. Due to the curse of dimensionality, classical and sampling approaches propagating parameter uncertainties to predictions are hardly feasible and insufficient. However, for experimental design and to discriminate between competing models, prediction and confidence bands are essential. To circumvent the hurdles of the former methods, an approach to calculate a profile likelihood on arbitrary observations for a specific time point has been introduced, which provides accurate confidence and prediction intervals for nonlinear models and is computationally feasible for high-dimensional models. In this article, reliable and smooth point-wise prediction and confidence bands to assess the model's uncertainty on the whole time-course are achieved via explicit integration with elaborate correction mechanisms. The corresponding system of ordinary differential equations is derived and tested on three established models for cellular signalling. An efficiency analysis is performed to illustrate the computational benefit compared with repeated profile likelihood calculations at multiple time points. The integration framework and the examples used in this article are provided with the software package Data2Dynamics, which is based on MATLAB and freely available at http://www.data2dynamics.org helge.hass@fdm.uni-freiburg.de Supplementary data are available at Bioinformatics online. © The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please e

Prediction of multi-wake problems using an improved Jensen wake model

DEFF Research Database (Denmark)

Tian, Linlin; Zhu, Wei Jun; Shen, Wen Zhong

2017-01-01

The improved analytical wake model named as 2D_k Jensen model (which was proposed to overcome some shortcomes in the classical Jensen wake model) is applied and validated in this work for wind turbine multi-wake predictions. Different from the original Jensen model, this newly developed 2D_k Jensen...... model uses a cosine shape instead of the top-hat shape for the velocity deficit in the wake, and the wake decay rate as a variable that is related to the ambient turbulence as well as the rotor generated turbulence. Coupled with four different multi-wake combination models, the 2D_k Jensen model...... is assessed through (1) simulating two wakes interaction under full wake and partial wake conditions and (2) predicting the power production in the Horns Rev wind farm for different wake sectors around two different wind directions. Through comparisons with field measurements, results from Large Eddy...

Evolution and application of a pseudo-multi-zone model for the prediction of NOx emissions from large-scale diesel engines at various operating conditions

International Nuclear Information System (INIS)

Savva, Nicholas S.; Hountalas, Dimitrios T.

2014-01-01

Highlights: • Development of a simplified simulation model for NO x formation during combustion. • Application of the proposed model on large-scale two and four-stroke diesel engines. • Experimental data from stationary and ship main and auxiliary engines were used. • The model captures the trend of NO x as engine power and fuel injection timing varies. • The model is recommended for research and practical use in maritime and power industry. - Abstract: Emissions regulations for heavy-duty diesel units used in maritime and power generation applications have become very strict the last years. Hence, the industry is enforced to limit specific gaseous and particulate emissions (NO x , SO x , CO x , PM and HC) depending on the regulations. Among numerous methods, simulation models are extensively used to support the development of techniques used for the control of emitted pollutants. This is very important for large-scale engines due to the extremely high cost of the experimental investigation resulting from the size of the engines and the test equipment involved. Beyond this, simulation models can also be used to support NO x monitoring, since on-board verification techniques are to become mandatory for the marine industry in the near future. Last but not least, simulation models can also be used for model-based control applications to support the operation of both in-cylinder and after-treatment techniques. Currently, the major controlled pollutant for both marine and stationary applications is NO x . For this reason, in the present work, authors focus on the development and application of a simplified NO x model with special emphasis on its ability to predict the effect of operating conditions on NO x for both two and four-stroke diesel engines. To accomplish this, an existing well validated simplified NO x model has been modified to enhance its physical background and applied on 16 different large-scale diesel engines utilizing 18 different sets of

Multi-fidelity uncertainty quantification in large-scale predictive simulations of turbulent flow

Science.gov (United States)

Geraci, Gianluca; Jofre-Cruanyes, Lluis; Iaccarino, Gianluca

2017-11-01

The performance characterization of complex engineering systems often relies on accurate, but computationally intensive numerical simulations. It is also well recognized that in order to obtain a reliable numerical prediction the propagation of uncertainties needs to be included. Therefore, Uncertainty Quantification (UQ) plays a fundamental role in building confidence in predictive science. Despite the great improvement in recent years, even the more advanced UQ algorithms are still limited to fairly simplified applications and only moderate parameter dimensionality. Moreover, in the case of extremely large dimensionality, sampling methods, i.e. Monte Carlo (MC) based approaches, appear to be the only viable alternative. In this talk we describe and compare a family of approaches which aim to accelerate the convergence of standard MC simulations. These methods are based on hierarchies of generalized numerical resolutions (multi-level) or model fidelities (multi-fidelity), and attempt to leverage the correlation between Low- and High-Fidelity (HF) models to obtain a more accurate statistical estimator without introducing additional HF realizations. The performance of these methods are assessed on an irradiated particle laden turbulent flow (PSAAP II solar energy receiver). This investigation was funded by the United States Department of Energy's (DoE) National Nuclear Security Administration (NNSA) under the Predicitive Science Academic Alliance Program (PSAAP) II at Stanford University.

Sensitivity analysis of predictive models with an automated adjoint generator

International Nuclear Information System (INIS)

Pin, F.G.; Oblow, E.M.

1987-01-01

The adjoint method is a well established sensitivity analysis methodology that is particularly efficient in large-scale modeling problems. The coefficients of sensitivity of a given response with respect to every parameter involved in the modeling code can be calculated from the solution of a single adjoint run of the code. Sensitivity coefficients provide a quantitative measure of the importance of the model data in calculating the final results. The major drawback of the adjoint method is the requirement for calculations of very large numbers of partial derivatives to set up the adjoint equations of the model. ADGEN is a software system that has been designed to eliminate this drawback and automatically implement the adjoint formulation in computer codes. The ADGEN system will be described and its use for improving performance assessments and predictive simulations will be discussed. 8 refs., 1 fig

Predictive Modeling in Race Walking

Directory of Open Access Journals (Sweden)

Krzysztof Wiktorowicz

2015-01-01

Full Text Available This paper presents the use of linear and nonlinear multivariable models as tools to support training process of race walkers. These models are calculated using data collected from race walkers’ training events and they are used to predict the result over a 3 km race based on training loads. The material consists of 122 training plans for 21 athletes. In order to choose the best model leave-one-out cross-validation method is used. The main contribution of the paper is to propose the nonlinear modifications for linear models in order to achieve smaller prediction error. It is shown that the best model is a modified LASSO regression with quadratic terms in the nonlinear part. This model has the smallest prediction error and simplified structure by eliminating some of the predictors.

Application of simplified models to CO2 migration and immobilization in large-scale geological systems

KAUST Repository

Gasda, Sarah E.

2012-07-01

Long-term stabilization of injected carbon dioxide (CO 2) is an essential component of risk management for geological carbon sequestration operations. However, migration and trapping phenomena are inherently complex, involving processes that act over multiple spatial and temporal scales. One example involves centimeter-scale density instabilities in the dissolved CO 2 region leading to large-scale convective mixing that can be a significant driver for CO 2 dissolution. Another example is the potentially important effect of capillary forces, in addition to buoyancy and viscous forces, on the evolution of mobile CO 2. Local capillary effects lead to a capillary transition zone, or capillary fringe, where both fluids are present in the mobile state. This small-scale effect may have a significant impact on large-scale plume migration as well as long-term residual and dissolution trapping. Computational models that can capture both large and small-scale effects are essential to predict the role of these processes on the long-term storage security of CO 2 sequestration operations. Conventional modeling tools are unable to resolve sufficiently all of these relevant processes when modeling CO 2 migration in large-scale geological systems. Herein, we present a vertically-integrated approach to CO 2 modeling that employs upscaled representations of these subgrid processes. We apply the model to the Johansen formation, a prospective site for sequestration of Norwegian CO 2 emissions, and explore the sensitivity of CO 2 migration and trapping to subscale physics. Model results show the relative importance of different physical processes in large-scale simulations. The ability of models such as this to capture the relevant physical processes at large spatial and temporal scales is important for prediction and analysis of CO 2 storage sites. © 2012 Elsevier Ltd.

Prediction of early summer rainfall over South China by a physical-empirical model

Science.gov (United States)

Yim, So-Young; Wang, Bin; Xing, Wen

2014-10-01

In early summer (May-June, MJ) the strongest rainfall belt of the northern hemisphere occurs over the East Asian (EA) subtropical front. During this period the South China (SC) rainfall reaches its annual peak and represents the maximum rainfall variability over EA. Hence we establish an SC rainfall index, which is the MJ mean precipitation averaged over 72 stations over SC (south of 28°N and east of 110°E) and represents superbly the leading empirical orthogonal function mode of MJ precipitation variability over EA. In order to predict SC rainfall, we established a physical-empirical model. Analysis of 34-year observations (1979-2012) reveals three physically consequential predictors. A plentiful SC rainfall is preceded in the previous winter by (a) a dipole sea surface temperature (SST) tendency in the Indo-Pacific warm pool, (b) a tripolar SST tendency in North Atlantic Ocean, and (c) a warming tendency in northern Asia. These precursors foreshadow enhanced Philippine Sea subtropical High and Okhotsk High in early summer, which are controlling factors for enhanced subtropical frontal rainfall. The physical empirical model built on these predictors achieves a cross-validated forecast correlation skill of 0.75 for 1979-2012. Surprisingly, this skill is substantially higher than four-dynamical models' ensemble prediction for 1979-2010 period (0.15). The results here suggest that the low prediction skill of current dynamical models is largely due to models' deficiency and the dynamical prediction has large room to improve.

Predictive modeling in e-mental health: A common language framework

Directory of Open Access Journals (Sweden)

Dennis Becker

2018-06-01

Full Text Available Recent developments in mobile technology, sensor devices, and artificial intelligence have created new opportunities for mental health care research. Enabled by large datasets collected in e-mental health research and practice, clinical researchers and members of the data mining community increasingly join forces to build predictive models for health monitoring, treatment selection, and treatment personalization. This paper aims to bridge the historical and conceptual gaps between the distant research domains involved in this new collaborative research by providing a conceptual model of common research goals. We first provide a brief overview of the data mining field and methods used for predictive modeling. Next, we propose to characterize predictive modeling research in mental health care on three dimensions: 1 time, relative to treatment (i.e., from screening to post-treatment relapse monitoring, 2 types of available data (e.g., questionnaire data, ecological momentary assessments, smartphone sensor data, and 3 type of clinical decision (i.e., whether data are used for screening purposes, treatment selection or treatment personalization. Building on these three dimensions, we introduce a framework that identifies four model types that can be used to classify existing and future research and applications. To illustrate this, we use the framework to classify and discuss published predictive modeling mental health research. Finally, in the discussion, we reflect on the next steps that are required to drive forward this promising new interdisciplinary field.

Using radar altimetry to update a large-scale hydrological model of the Brahmaputra river basin

DEFF Research Database (Denmark)

Finsen, F.; Milzow, Christian; Smith, R.

2014-01-01

Measurements of river and lake water levels from space-borne radar altimeters (past missions include ERS, Envisat, Jason, Topex) are useful for calibration and validation of large-scale hydrological models in poorly gauged river basins. Altimetry data availability over the downstream reaches...... of the Brahmaputra is excellent (17 high-quality virtual stations from ERS-2, 6 from Topex and 10 from Envisat are available for the Brahmaputra). In this study, altimetry data are used to update a large-scale Budyko-type hydrological model of the Brahmaputra river basin in real time. Altimetry measurements...... improved model performance considerably. The Nash-Sutcliffe model efficiency increased from 0.77 to 0.83. Real-time river basin modelling using radar altimetry has the potential to improve the predictive capability of large-scale hydrological models elsewhere on the planet....

Modeling heterogeneous (co)variances from adjacent-SNP groups improves genomic prediction for milk protein composition traits

DEFF Research Database (Denmark)

Gebreyesus, Grum; Lund, Mogens Sandø; Buitenhuis, Albert Johannes

2017-01-01

Accurate genomic prediction requires a large reference population, which is problematic for traits that are expensive to measure. Traits related to milk protein composition are not routinely recorded due to costly procedures and are considered to be controlled by a few quantitative trait loci...... of large effect. The amount of variation explained may vary between regions leading to heterogeneous (co)variance patterns across the genome. Genomic prediction models that can efficiently take such heterogeneity of (co)variances into account can result in improved prediction reliability. In this study, we...... developed and implemented novel univariate and bivariate Bayesian prediction models, based on estimates of heterogeneous (co)variances for genome segments (BayesAS). Available data consisted of milk protein composition traits measured on cows and de-regressed proofs of total protein yield derived for bulls...

Adding propensity scores to pure prediction models fails to improve predictive performance

Directory of Open Access Journals (Sweden)

Amy S. Nowacki

2013-08-01

Full Text Available Background. Propensity score usage seems to be growing in popularity leading researchers to question the possible role of propensity scores in prediction modeling, despite the lack of a theoretical rationale. It is suspected that such requests are due to the lack of differentiation regarding the goals of predictive modeling versus causal inference modeling. Therefore, the purpose of this study is to formally examine the effect of propensity scores on predictive performance. Our hypothesis is that a multivariable regression model that adjusts for all covariates will perform as well as or better than those models utilizing propensity scores with respect to model discrimination and calibration.Methods. The most commonly encountered statistical scenarios for medical prediction (logistic and proportional hazards regression were used to investigate this research question. Random cross-validation was performed 500 times to correct for optimism. The multivariable regression models adjusting for all covariates were compared with models that included adjustment for or weighting with the propensity scores. The methods were compared based on three predictive performance measures: (1 concordance indices; (2 Brier scores; and (3 calibration curves.Results. Multivariable models adjusting for all covariates had the highest average concordance index, the lowest average Brier score, and the best calibration. Propensity score adjustment and inverse probability weighting models without adjustment for all covariates performed worse than full models and failed to improve predictive performance with full covariate adjustment.Conclusion. Propensity score techniques did not improve prediction performance measures beyond multivariable adjustment. Propensity scores are not recommended if the analytical goal is pure prediction modeling.

A prediction model for assessing residential radon concentration in Switzerland

International Nuclear Information System (INIS)

Hauri, Dimitri D.; Huss, Anke; Zimmermann, Frank; Kuehni, Claudia E.; Röösli, Martin

2012-01-01

Indoor radon is regularly measured in Switzerland. However, a nationwide model to predict residential radon levels has not been developed. The aim of this study was to develop a prediction model to assess indoor radon concentrations in Switzerland. The model was based on 44,631 measurements from the nationwide Swiss radon database collected between 1994 and 2004. Of these, 80% randomly selected measurements were used for model development and the remaining 20% for an independent model validation. A multivariable log-linear regression model was fitted and relevant predictors selected according to evidence from the literature, the adjusted R², the Akaike's information criterion (AIC), and the Bayesian information criterion (BIC). The prediction model was evaluated by calculating Spearman rank correlation between measured and predicted values. Additionally, the predicted values were categorised into three categories (50th, 50th–90th and 90th percentile) and compared with measured categories using a weighted Kappa statistic. The most relevant predictors for indoor radon levels were tectonic units and year of construction of the building, followed by soil texture, degree of urbanisation, floor of the building where the measurement was taken and housing type (P-values <0.001 for all). Mean predicted radon values (geometric mean) were 66 Bq/m³ (interquartile range 40–111 Bq/m³) in the lowest exposure category, 126 Bq/m³ (69–215 Bq/m³) in the medium category, and 219 Bq/m³ (108–427 Bq/m³) in the highest category. Spearman correlation between predictions and measurements was 0.45 (95%-CI: 0.44; 0.46) for the development dataset and 0.44 (95%-CI: 0.42; 0.46) for the validation dataset. Kappa coefficients were 0.31 for the development and 0.30 for the validation dataset, respectively. The model explained 20% overall variability (adjusted R²). In conclusion, this residential radon prediction model, based on a large number of measurements, was demonstrated to be

Model-free and model-based reward prediction errors in EEG.

Science.gov (United States)

Sambrook, Thomas D; Hardwick, Ben; Wills, Andy J; Goslin, Jeremy

2018-05-24

Learning theorists posit two reinforcement learning systems: model-free and model-based. Model-based learning incorporates knowledge about structure and contingencies in the world to assign candidate actions with an expected value. Model-free learning is ignorant of the world's structure; instead, actions hold a value based on prior reinforcement, with this value updated by expectancy violation in the form of a reward prediction error. Because they use such different learning mechanisms, it has been previously assumed that model-based and model-free learning are computationally dissociated in the brain. However, recent fMRI evidence suggests that the brain may compute reward prediction errors to both model-free and model-based estimates of value, signalling the possibility that these systems interact. Because of its poor temporal resolution, fMRI risks confounding reward prediction errors with other feedback-related neural activity. In the present study, EEG was used to show the presence of both model-based and model-free reward prediction errors and their place in a temporal sequence of events including state prediction errors and action value updates. This demonstration of model-based prediction errors questions a long-held assumption that model-free and model-based learning are dissociated in the brain. Copyright © 2018 Elsevier Inc. All rights reserved.

Pressure fluctuation prediction of a model pump turbine at no load opening by a nonlinear k-ε turbulence model

International Nuclear Information System (INIS)

Liu, J T; Zuo, Z G; Liu, S H; Wu, Y L

2014-01-01

In this paper, a new nonlinear k-ε turbulence model based on RNG k-ε turbulence model and Wilcox's k-ω turbulence model was proposed to simulate the unsteady flow and to predict the pressure fluctuation through a model pump turbine for engineering application. Calculations on a curved rectangular duct proved that the nonlinear k-ε turbulence model is applicable for high pressure gradient flows and large curvature flows. The numerically predicted relative pressure amplitude (peak to peak) in time domain to the pump turbine head at no load condition is very close to the experimental data. It is indicated that the prediction of the pressure fluctuation is valid by the present nonlinear k-ε method. The high pressure fluctuation in this area is the main issue for pump turbine design, especially at high head condition

Model-based predictive control scheme for cost optimization and balancing services for supermarket refrigeration Systems

NARCIS (Netherlands)

Weerts, H.H.M.; Shafiei, S.E.; Stoustrup, J.; Izadi-Zamanabadi, R.; Boje, E.; Xia, X.

2014-01-01

A new formulation of model predictive control for supermarket refrigeration systems is proposed to facilitate the regulatory power services as well as energy cost optimization of such systems in the smart grid. Nonlinear dynamics existed in large-scale refrigeration plants challenges the predictive

Nonlinear chaotic model for predicting storm surges

Directory of Open Access Journals (Sweden)

M. Siek

2010-09-01

Full Text Available This paper addresses the use of the methods of nonlinear dynamics and chaos theory for building a predictive chaotic model from time series. The chaotic model predictions are made by the adaptive local models based on the dynamical neighbors found in the reconstructed phase space of the observables. We implemented the univariate and multivariate chaotic models with direct and multi-steps prediction techniques and optimized these models using an exhaustive search method. The built models were tested for predicting storm surge dynamics for different stormy conditions in the North Sea, and are compared to neural network models. The results show that the chaotic models can generally provide reliable and accurate short-term storm surge predictions.

Global Bedload Flux Modeling and Analysis in Large Rivers

Science.gov (United States)

Islam, M. T.; Cohen, S.; Syvitski, J. P.

2017-12-01

Proper sediment transport quantification has long been an area of interest for both scientists and engineers in the fields of geomorphology, and management of rivers and coastal waters. Bedload flux is important for monitoring water quality and for sustainable development of coastal and marine bioservices. Bedload measurements, especially for large rivers, is extremely scarce across time, and many rivers have never been monitored. Bedload measurements in rivers, is particularly acute in developing countries where changes in sediment yields is high. The paucity of bedload measurements is the result of 1) the nature of the problem (large spatial and temporal uncertainties), and 2) field costs including the time-consuming nature of the measurement procedures (repeated bedform migration tracking, bedload samplers). Here we present a first of its kind methodology for calculating bedload in large global rivers (basins are >1,000 km. Evaluation of model skill is based on 113 bedload measurements. The model predictions are compared with an empirical model developed from the observational dataset in an attempt to evaluate the differences between a physically-based numerical model and a lumped relationship between bedload flux and fluvial and basin parameters (e.g., discharge, drainage area, lithology). The initial study success opens up various applications to global fluvial geomorphology (e.g. including the relationship between suspended sediment (wash load) and bedload). Simulated results with known uncertainties offers a new research product as a valuable resource for the whole scientific community.

Machine learning models in breast cancer survival prediction.

Science.gov (United States)

Montazeri, Mitra; Montazeri, Mohadeseh; Montazeri, Mahdieh; Beigzadeh, Amin

2016-01-01

Breast cancer is one of the most common cancers with a high mortality rate among women. With the early diagnosis of breast cancer survival will increase from 56% to more than 86%. Therefore, an accurate and reliable system is necessary for the early diagnosis of this cancer. The proposed model is the combination of rules and different machine learning techniques. Machine learning models can help physicians to reduce the number of false decisions. They try to exploit patterns and relationships among a large number of cases and predict the outcome of a disease using historical cases stored in datasets. The objective of this study is to propose a rule-based classification method with machine learning techniques for the prediction of different types of Breast cancer survival. We use a dataset with eight attributes that include the records of 900 patients in which 876 patients (97.3%) and 24 (2.7%) patients were females and males respectively. Naive Bayes (NB), Trees Random Forest (TRF), 1-Nearest Neighbor (1NN), AdaBoost (AD), Support Vector Machine (SVM), RBF Network (RBFN), and Multilayer Perceptron (MLP) machine learning techniques with 10-cross fold technique were used with the proposed model for the prediction of breast cancer survival. The performance of machine learning techniques were evaluated with accuracy, precision, sensitivity, specificity, and area under ROC curve. Out of 900 patients, 803 patients and 97 patients were alive and dead, respectively. In this study, Trees Random Forest (TRF) technique showed better results in comparison to other techniques (NB, 1NN, AD, SVM and RBFN, MLP). The accuracy, sensitivity and the area under ROC curve of TRF are 96%, 96%, 93%, respectively. However, 1NN machine learning technique provided poor performance (accuracy 91%, sensitivity 91% and area under ROC curve 78%). This study demonstrates that Trees Random Forest model (TRF) which is a rule-based classification model was the best model with the highest level of

Micromechanics-based damage model for failure prediction in cold forming

Energy Technology Data Exchange (ETDEWEB)

Lu, X.Z.; Chan, L.C., E-mail: lc.chan@polyu.edu.hk

2017-04-06

The purpose of this study was to develop a micromechanics-based damage (micro-damage) model that was concerned with the evolution of micro-voids for failure prediction in cold forming. Typical stainless steel SS316L was selected as the specimen material, and the nonlinear isotropic hardening rule was extended to describe the large deformation of the specimen undergoing cold forming. A micro-focus high-resolution X-ray computed tomography (CT) system was employed to trace and measure the micro-voids inside the specimen directly. Three-dimensional (3D) representative volume element (RVE) models with different sizes and spatial locations were reconstructed from the processed CT images of the specimen, and the average size and volume fraction of micro-voids (VFMV) for the specimen were determined via statistical analysis. Subsequently, the micro-damage model was compiled as a user-defined material subroutine into the finite element (FE) package ABAQUS. The stress-strain responses and damage evolutions of SS316L specimens under tensile and compressive deformations at different strain rates were predicted and further verified experimentally. It was concluded that the proposed micro-damage model is convincing for failure prediction in cold forming of the SS316L material.

Prediction of Canopy Heights over a Large Region Using Heterogeneous Lidar Datasets: Efficacy and Challenges

Directory of Open Access Journals (Sweden)

Ranjith Gopalakrishnan

2015-08-01

Full Text Available Generating accurate and unbiased wall-to-wall canopy height maps from airborne lidar data for large regions is useful to forest scientists and natural resource managers. However, mapping large areas often involves using lidar data from different projects, with varying acquisition parameters. In this work, we address the important question of whether one can accurately model canopy heights over large areas of the Southeastern US using a very heterogeneous dataset of small-footprint, discrete-return airborne lidar data (with 76 separate lidar projects. A unique aspect of this effort is the use of nationally uniform and extensive field data (~1800 forested plots from the Forest Inventory and Analysis (FIA program of the US Forest Service. Preliminary results are quite promising: Over all lidar projects, we observe a good correlation between the 85th percentile of lidar heights and field-measured height (r = 0.85. We construct a linear regression model to predict subplot-level dominant tree heights from distributional lidar metrics (R2 = 0.74, RMSE = 3.0 m, n = 1755. We also identify and quantify the importance of several factors (like heterogeneity of vegetation, point density, the predominance of hardwoods or softwoods, the average height of the forest stand, slope of the plot, and average scan angle of lidar acquisition that influence the efficacy of predicting canopy heights from lidar data. For example, a subset of plots (coefficient of variation of vegetation heights <0.2 significantly reduces the RMSE of our model from 3.0–2.4 m (~20% reduction. We conclude that when all these elements are factored into consideration, combining data from disparate lidar projects does not preclude robust estimation of canopy heights.

Nonlinear dynamical modeling and prediction of the terrestrial magnetospheric activity

International Nuclear Information System (INIS)

Vassiliadis, D.

1992-01-01

The irregular activity of the magnetosphere results from its complex internal dynamics as well as the external influence of the solar wind. The dominating self-organization of the magnetospheric plasma gives rise to repetitive, large-scale coherent behavior manifested in phenomena such as the magnetic substorm. Based on the nonlinearity of the global dynamics this dissertation examines the magnetosphere as a nonlinear dynamical system using time series analysis techniques. Initially the magnetospheric activity is modeled in terms of an autonomous system. A dimension study shows that its observed time series is self-similar, but the correlation dimension is high. The implication of a large number of degrees of freedom is confirmed by other state space techniques such as Poincare sections and search for unstable periodic orbits. At the same time a stability study of the time series in terms of Lyapunov exponents suggests that the series is not chaotic. The absence of deterministic chaos is supported by the low predictive capability of the autonomous model. Rather than chaos, it is an external input which is largely responsible for the irregularity of the magnetospheric activity. In fact, the external driving is so strong that the above state space techniques give results for magnetospheric and solar wind time series that are at least qualitatively similar. Therefore the solar wind input has to be included in a low-dimensional nonautonomous model. Indeed it is shown that such a model can reproduce the observed magnetospheric behavior up to 80-90 percent. The characteristic coefficients of the model show little variation depending on the external disturbance. The impulse response is consistent with earlier results of linear prediction filters. The model can be easily extended to contain nonlinear features of the magnetospheric activity and in particular the loading-unloading behavior of substorms

Validated predictive modelling of the environmental resistome.

Science.gov (United States)

Amos, Gregory C A; Gozzard, Emma; Carter, Charlotte E; Mead, Andrew; Bowes, Mike J; Hawkey, Peter M; Zhang, Lihong; Singer, Andrew C; Gaze, William H; Wellington, Elizabeth M H

2015-06-01

Multi-drug-resistant bacteria pose a significant threat to public health. The role of the environment in the overall rise in antibiotic-resistant infections and risk to humans is largely unknown. This study aimed to evaluate drivers of antibiotic-resistance levels across the River Thames catchment, model key biotic, spatial and chemical variables and produce predictive models for future risk assessment. Sediment samples from 13 sites across the River Thames basin were taken at four time points across 2011 and 2012. Samples were analysed for class 1 integron prevalence and enumeration of third-generation cephalosporin-resistant bacteria. Class 1 integron prevalence was validated as a molecular marker of antibiotic resistance; levels of resistance showed significant geospatial and temporal variation. The main explanatory variables of resistance levels at each sample site were the number, proximity, size and type of surrounding wastewater-treatment plants. Model 1 revealed treatment plants accounted for 49.5% of the variance in resistance levels. Other contributing factors were extent of different surrounding land cover types (for example, Neutral Grassland), temporal patterns and prior rainfall; when modelling all variables the resulting model (Model 2) could explain 82.9% of variations in resistance levels in the whole catchment. Chemical analyses correlated with key indicators of treatment plant effluent and a model (Model 3) was generated based on water quality parameters (contaminant and macro- and micro-nutrient levels). Model 2 was beta tested on independent sites and explained over 78% of the variation in integron prevalence showing a significant predictive ability. We believe all models in this study are highly useful tools for informing and prioritising mitigation strategies to reduce the environmental resistome.

Large-Angle CMB Suppression and Polarisation Predictions

CERN Document Server

Copi, C.J.; Schwarz, D.J.; Starkman, G.D.

2013-01-01

The anomalous lack of large angle temperature correlations has been a surprising feature of the CMB since first observed by COBE-DMR and subsequently confirmed and strengthened by WMAP. This anomaly may point to the need for modifications of the standard model of cosmology or may show that our Universe is a rare statistical fluctuation within that model. Further observations of the temperature auto-correlation function will not elucidate the issue; sufficiently high precision statistical observations already exist. Instead, alternative probes are required. In this work we explore the expectations for forthcoming polarisation observations. We define a prescription to test the hypothesis that the large-angle CMB temperature perturbations in our Universe represent a rare statistical fluctuation within the standard cosmological model. These tests are based on the temperature-Q Stokes parameter correlation. Unfortunately these tests cannot be expected to be definitive. However, we do show that if this TQ-correlati...

Determining the validity of exposure models for environmental epidemiology : predicting electromagnetic fields from mobile phone base stations

NARCIS (Netherlands)

Beekhuizen, Johan|info:eu-repo/dai/nl/34472641X

2014-01-01

One of the key challenges in environmental epidemiology is the exposure assessment of large populations. Spatial exposure models have been developed that predict exposure to the pollutant of interest for large study sizes. However, the validity of these exposure models is often unknown. In this

Predicting watershed sediment yields after wildland fire with the InVEST sediment retention model at large geographic extent in the western USA: accuracy and uncertainties

Science.gov (United States)

Sankey, J. B.; Kreitler, J.; McVay, J.; Hawbaker, T. J.; Vaillant, N.; Lowe, S. E.

2014-12-01

Wildland fire is a primary threat to watersheds that can impact water supply through increased sedimentation, water quality decline, and change the timing and amount of runoff leading to increased risk from flood and sediment natural hazards. It is of great societal importance in the western USA and throughout the world to improve understanding of how changing fire frequency, extent, and location, in conjunction with fuel treatments will affect watersheds and the ecosystem services they supply to communities. In this work we assess the utility of the InVEST Sediment Retention Model to accurately characterize vulnerability of burned watersheds to erosion and sedimentation. The InVEST tools are GIS-based implementations of common process models, engineered for high-end computing to allow the faster simulation of larger landscapes and incorporation into decision-making. The InVEST Sediment Retention Model is based on common soil erosion models (e.g., RUSLE -Revised Universal Soil Loss Equation) and determines which areas of the landscape contribute the greatest sediment loads to a hydrological network and conversely evaluate the ecosystem service of sediment retention on a watershed basis. We evaluate the accuracy and uncertainties for InVEST predictions of increased sedimentation after fire, using measured post-fire sedimentation rates available for many watersheds in different rainfall regimes throughout the western USA from an existing, large USGS database of post-fire sediment yield [synthesized in Moody J, Martin D (2009) Synthesis of sediment yields after wildland fire in different rainfall regimes in the western United States. International Journal of Wildland Fire 18: 96-115]. The ultimate goal of this work is to calibrate and implement the model to accurately predict variability in post-fire sediment yield as a function of future landscape heterogeneity predicted by wildfire simulations, and future landscape fuel treatment scenarios, within watersheds.

Extracting falsifiable predictions from sloppy models.

Science.gov (United States)

Gutenkunst, Ryan N; Casey, Fergal P; Waterfall, Joshua J; Myers, Christopher R; Sethna, James P

2007-12-01

Successful predictions are among the most compelling validations of any model. Extracting falsifiable predictions from nonlinear multiparameter models is complicated by the fact that such models are commonly sloppy, possessing sensitivities to different parameter combinations that range over many decades. Here we discuss how sloppiness affects the sorts of data that best constrain model predictions, makes linear uncertainty approximations dangerous, and introduces computational difficulties in Monte-Carlo uncertainty analysis. We also present a useful test problem and suggest refinements to the standards by which models are communicated.

A Bayesian antedependence model for whole genome prediction.

Science.gov (United States)

Yang, Wenzhao; Tempelman, Robert J

2012-04-01

Hierarchical mixed effects models have been demonstrated to be powerful for predicting genomic merit of livestock and plants, on the basis of high-density single-nucleotide polymorphism (SNP) marker panels, and their use is being increasingly advocated for genomic predictions in human health. Two particularly popular approaches, labeled BayesA and BayesB, are based on specifying all SNP-associated effects to be independent of each other. BayesB extends BayesA by allowing a large proportion of SNP markers to be associated with null effects. We further extend these two models to specify SNP effects as being spatially correlated due to the chromosomally proximal effects of causal variants. These two models, that we respectively dub as ante-BayesA and ante-BayesB, are based on a first-order nonstationary antedependence specification between SNP effects. In a simulation study involving 20 replicate data sets, each analyzed at six different SNP marker densities with average LD levels ranging from r(2) = 0.15 to 0.31, the antedependence methods had significantly (P 0. 24) with differences exceeding 3%. A cross-validation study was also conducted on the heterogeneous stock mice data resource (http://mus.well.ox.ac.uk/mouse/HS/) using 6-week body weights as the phenotype. The antedependence methods increased cross-validation prediction accuracies by up to 3.6% compared to their classical counterparts (P benchmark data sets and demonstrated that the antedependence methods were more accurate than their classical counterparts for genomic predictions, even for individuals several generations beyond the training data.

Covariance approximation for large multivariate spatial data sets with an application to multiple climate model errors

KAUST Repository

Sang, Huiyan

2011-12-01

This paper investigates the cross-correlations across multiple climate model errors. We build a Bayesian hierarchical model that accounts for the spatial dependence of individual models as well as cross-covariances across different climate models. Our method allows for a nonseparable and nonstationary cross-covariance structure. We also present a covariance approximation approach to facilitate the computation in the modeling and analysis of very large multivariate spatial data sets. The covariance approximation consists of two parts: a reduced-rank part to capture the large-scale spatial dependence, and a sparse covariance matrix to correct the small-scale dependence error induced by the reduced rank approximation. We pay special attention to the case that the second part of the approximation has a block-diagonal structure. Simulation results of model fitting and prediction show substantial improvement of the proposed approximation over the predictive process approximation and the independent blocks analysis. We then apply our computational approach to the joint statistical modeling of multiple climate model errors. © 2012 Institute of Mathematical Statistics.

Modelling for the Stripa site characterization and validation drift inflow: prediction of flow through fractured rock

International Nuclear Information System (INIS)

Herbert, A.; Gale, J.; MacLeod, R.; Lanyon, G.

1991-12-01

We present our approach to predicting flow through a fractured rock site; the site characterization and validation region in the Stripa mine. Our approach is based on discrete fracture network modelling using the NAPSAC computer code. We describe the conceptual models and assumptions that we have used to interpret the geometry and flow properties of the fracture networks, from measurements at the site. These are used to investigate large scale properties of the network and we show that for flows on scales larger than about 10 m, porous medium approximation should be used. The porous medium groundwater flow code CFEST is used to predict the large scale flows through the mine and the SCV region. This, in turn, is used to provide boundary conditions for more detailed models, which predict the details of flow, using a discrete fracture network model, on scales of less than 10 m. We conclude that a fracture network approach is feasible and that it provides a better understanding of details of flow than conventional porous medium approaches and a quantification of the uncertainty associated with predictive flow modelling characterised from field measurement in fractured rock. (au)

Fixed recurrence and slip models better predict earthquake behavior than the time- and slip-predictable models 1: repeating earthquakes

Science.gov (United States)

Rubinstein, Justin L.; Ellsworth, William L.; Chen, Kate Huihsuan; Uchida, Naoki

2012-01-01

The behavior of individual events in repeating earthquake sequences in California, Taiwan and Japan is better predicted by a model with fixed inter-event time or fixed slip than it is by the time- and slip-predictable models for earthquake occurrence. Given that repeating earthquakes are highly regular in both inter-event time and seismic moment, the time- and slip-predictable models seem ideally suited to explain their behavior. Taken together with evidence from the companion manuscript that shows similar results for laboratory experiments we conclude that the short-term predictions of the time- and slip-predictable models should be rejected in favor of earthquake models that assume either fixed slip or fixed recurrence interval. This implies that the elastic rebound model underlying the time- and slip-predictable models offers no additional value in describing earthquake behavior in an event-to-event sense, but its value in a long-term sense cannot be determined. These models likely fail because they rely on assumptions that oversimplify the earthquake cycle. We note that the time and slip of these events is predicted quite well by fixed slip and fixed recurrence models, so in some sense they are time- and slip-predictable. While fixed recurrence and slip models better predict repeating earthquake behavior than the time- and slip-predictable models, we observe a correlation between slip and the preceding recurrence time for many repeating earthquake sequences in Parkfield, California. This correlation is not found in other regions, and the sequences with the correlative slip-predictable behavior are not distinguishable from nearby earthquake sequences that do not exhibit this behavior.

Performance modeling of hybrid MPI/OpenMP scientific applications on large-scale multicore supercomputers

KAUST Repository

Wu, Xingfu; Taylor, Valerie

2013-01-01

In this paper, we present a performance modeling framework based on memory bandwidth contention time and a parameterized communication model to predict the performance of OpenMP, MPI and hybrid applications with weak scaling on three large-scale multicore supercomputers: IBM POWER4, POWER5+ and BlueGene/P, and analyze the performance of these MPI, OpenMP and hybrid applications. We use STREAM memory benchmarks and Intel's MPI benchmarks to provide initial performance analysis and model validation of MPI and OpenMP applications on these multicore supercomputers because the measured sustained memory bandwidth can provide insight into the memory bandwidth that a system should sustain on scientific applications with the same amount of workload per core. In addition to using these benchmarks, we also use a weak-scaling hybrid MPI/OpenMP large-scale scientific application: Gyrokinetic Toroidal Code (GTC) in magnetic fusion to validate our performance model of the hybrid application on these multicore supercomputers. The validation results for our performance modeling method show less than 7.77% error rate in predicting the performance of hybrid MPI/OpenMP GTC on up to 512 cores on these multicore supercomputers. © 2013 Elsevier Inc.

Performance modeling of hybrid MPI/OpenMP scientific applications on large-scale multicore supercomputers

KAUST Repository

Wu, Xingfu

2013-12-01

In this paper, we present a performance modeling framework based on memory bandwidth contention time and a parameterized communication model to predict the performance of OpenMP, MPI and hybrid applications with weak scaling on three large-scale multicore supercomputers: IBM POWER4, POWER5+ and BlueGene/P, and analyze the performance of these MPI, OpenMP and hybrid applications. We use STREAM memory benchmarks and Intel\\'s MPI benchmarks to provide initial performance analysis and model validation of MPI and OpenMP applications on these multicore supercomputers because the measured sustained memory bandwidth can provide insight into the memory bandwidth that a system should sustain on scientific applications with the same amount of workload per core. In addition to using these benchmarks, we also use a weak-scaling hybrid MPI/OpenMP large-scale scientific application: Gyrokinetic Toroidal Code (GTC) in magnetic fusion to validate our performance model of the hybrid application on these multicore supercomputers. The validation results for our performance modeling method show less than 7.77% error rate in predicting the performance of hybrid MPI/OpenMP GTC on up to 512 cores on these multicore supercomputers. © 2013 Elsevier Inc.

Prediction of Coal Face Gas Concentration by Multi-Scale Selective Ensemble Hybrid Modeling

Directory of Open Access Journals (Sweden)

WU Xiang

2014-06-01

Full Text Available A selective ensemble hybrid modeling prediction method based on wavelet transformation is proposed to improve the fitting and generalization capability of the existing prediction models of the coal face gas concentration, which has a strong stochastic volatility. Mallat algorithm was employed for the multi-scale decomposition and single-scale reconstruction of the gas concentration time series. Then, it predicted every subsequence by sparsely weighted multi unstable ELM(extreme learning machine predictor within method SERELM(sparse ensemble regressors of ELM. At last, it superimposed the predicted values of these models to obtain the predicted values of the original sequence. The proposed method takes advantage of characteristics of multi scale analysis of wavelet transformation, accuracy and fast characteristics of ELM prediction and the generalization ability of L1 regularized selective ensemble learning method. The results show that the forecast accuracy has large increase by using the proposed method. The average relative error is 0.65%, the maximum relative error is 4.16% and the probability of relative error less than 1% reaches 0.785.

Numerical predictions of particle dispersed two-phase flows, using the LSD and SSF models

International Nuclear Information System (INIS)

Avila, R.; Cervantes de Gortari, J.; Universidad Nacional Autonoma de Mexico, Mexico City. Facultad de Ingenieria)

1988-01-01

A modified version of a numerical scheme which is suitable to predict parabolic dispersed two-phase flow, is presented. The original version of this scheme was used to predict the test cases discussed during the 3rd workshop on TPF predictions in Belgrade, 1986. In this paper, two particle dispersion models are included which use the Lagrangian approach predicting test case 1 and 3 of the 4th workshop. For the prediction of test case 1 the Lagrangian Stochastic Deterministic model (LSD) is used providing acceptable good results of mean and turbulent quantities for both solid and gas phases; however, the computed void fraction distribution is not in agreement with the measurements at locations away from the inlet, especially near the walls. Test case 3 is predicted using both the LSD and the Stochastic Separated Flow (SSF) models. It was found that the effects of turbulence modulation are large when the LSD model is used, whereas the particles have a negligible influence on the continuous phase if the SSF model is utilized for the computations. Predictions of gas phase properties based on both models agree well with measurements; however, the agreement between calculated and measured solid phase properties is less satisfactory. (orig.)

EFFICIENT PREDICTIVE MODELLING FOR ARCHAEOLOGICAL RESEARCH

OpenAIRE

Balla, A.; Pavlogeorgatos, G.; Tsiafakis, D.; Pavlidis, G.

2014-01-01

The study presents a general methodology for designing, developing and implementing predictive modelling for identifying areas of archaeological interest. The methodology is based on documented archaeological data and geographical factors, geospatial analysis and predictive modelling, and has been applied to the identification of possible Macedonian tombs’ locations in Northern Greece. The model was tested extensively and the results were validated using a commonly used predictive gain, which...

Mathematical formulation to predict the harmonics of the superconducting Large Hadron Collider magnets

Directory of Open Access Journals (Sweden)

Nicholas Sammut

2006-01-01

Full Text Available CERN is currently assembling the LHC (Large Hadron Collider that will accelerate and bring in collision 7 TeV protons for high energy physics. Such a superconducting magnet-based accelerator can be controlled only when the field errors of production and installation of all magnetic elements are known to the required accuracy. The ideal way to compensate the field errors obviously is to have direct diagnostics on the beam. For the LHC, however, a system solely based on beam feedback may be too demanding. The present baseline for the LHC control system hence requires an accurate forecast of the magnetic field and the multipole field errors to reduce the burden on the beam-based feedback. The field model is the core of this magnetic prediction system, that we call the field description for the LHC (FIDEL. The model will provide the forecast of the magnetic field at a given time, magnet operating current, magnet ramp rate, magnet temperature, and magnet powering history. The model is based on the identification and physical decomposition of the effects that contribute to the total field in the magnet aperture of the LHC dipoles. Each effect is quantified using data obtained from series measurements, and modeled theoretically or empirically depending on the complexity of the physical phenomena involved. This paper presents the developments of the new finely tuned magnetic field model and, using the data accumulated through series tests to date, evaluates its accuracy and predictive capabilities over a sector of the machine.

Applied Distributed Model Predictive Control for Energy Efficient Buildings and Ramp Metering

Science.gov (United States)

Koehler, Sarah Muraoka

Industrial large-scale control problems present an interesting algorithmic design challenge. A number of controllers must cooperate in real-time on a network of embedded hardware with limited computing power in order to maximize system efficiency while respecting constraints and despite communication delays. Model predictive control (MPC) can automatically synthesize a centralized controller which optimizes an objective function subject to a system model, constraints, and predictions of disturbance. Unfortunately, the computations required by model predictive controllers for large-scale systems often limit its industrial implementation only to medium-scale slow processes. Distributed model predictive control (DMPC) enters the picture as a way to decentralize a large-scale model predictive control problem. The main idea of DMPC is to split the computations required by the MPC problem amongst distributed processors that can compute in parallel and communicate iteratively to find a solution. Some popularly proposed solutions are distributed optimization algorithms such as dual decomposition and the alternating direction method of multipliers (ADMM). However, these algorithms ignore two practical challenges: substantial communication delays present in control systems and also problem non-convexity. This thesis presents two novel and practically effective DMPC algorithms. The first DMPC algorithm is based on a primal-dual active-set method which achieves fast convergence, making it suitable for large-scale control applications which have a large communication delay across its communication network. In particular, this algorithm is suited for MPC problems with a quadratic cost, linear dynamics, forecasted demand, and box constraints. We measure the performance of this algorithm and show that it significantly outperforms both dual decomposition and ADMM in the presence of communication delay. The second DMPC algorithm is based on an inexact interior point method which is

Recurrent and Dynamic Models for Predicting Streaming Video Quality of Experience.

Science.gov (United States)

Bampis, Christos G; Li, Zhi; Katsavounidis, Ioannis; Bovik, Alan C

2018-07-01

Streaming video services represent a very large fraction of global bandwidth consumption. Due to the exploding demands of mobile video streaming services, coupled with limited bandwidth availability, video streams are often transmitted through unreliable, low-bandwidth networks. This unavoidably leads to two types of major streaming-related impairments: compression artifacts and/or rebuffering events. In streaming video applications, the end-user is a human observer; hence being able to predict the subjective Quality of Experience (QoE) associated with streamed videos could lead to the creation of perceptually optimized resource allocation strategies driving higher quality video streaming services. We propose a variety of recurrent dynamic neural networks that conduct continuous-time subjective QoE prediction. By formulating the problem as one of time-series forecasting, we train a variety of recurrent neural networks and non-linear autoregressive models to predict QoE using several recently developed subjective QoE databases. These models combine multiple, diverse neural network inputs, such as predicted video quality scores, rebuffering measurements, and data related to memory and its effects on human behavioral responses, using them to predict QoE on video streams impaired by both compression artifacts and rebuffering events. Instead of finding a single time-series prediction model, we propose and evaluate ways of aggregating different models into a forecasting ensemble that delivers improved results with reduced forecasting variance. We also deploy appropriate new evaluation metrics for comparing time-series predictions in streaming applications. Our experimental results demonstrate improved prediction performance that approaches human performance. An implementation of this work can be found at https://github.com/christosbampis/NARX_QoE_release.

Evaluation of Penalized and Nonpenalized Methods for Disease Prediction with Large-Scale Genetic Data

Directory of Open Access Journals (Sweden)

Sungho Won

2015-01-01

Full Text Available Owing to recent improvement of genotyping technology, large-scale genetic data can be utilized to identify disease susceptibility loci and this successful finding has substantially improved our understanding of complex diseases. However, in spite of these successes, most of the genetic effects for many complex diseases were found to be very small, which have been a big hurdle to build disease prediction model. Recently, many statistical methods based on penalized regressions have been proposed to tackle the so-called “large P and small N” problem. Penalized regressions including least absolute selection and shrinkage operator (LASSO and ridge regression limit the space of parameters, and this constraint enables the estimation of effects for very large number of SNPs. Various extensions have been suggested, and, in this report, we compare their accuracy by applying them to several complex diseases. Our results show that penalized regressions are usually robust and provide better accuracy than the existing methods for at least diseases under consideration.

Risk assessment and remedial policy evaluation using predictive modeling

International Nuclear Information System (INIS)

Linkov, L.; Schell, W.R.

1996-01-01

As a result of nuclear industry operation and accidents, large areas of natural ecosystems have been contaminated by radionuclides and toxic metals. Extensive societal pressure has been exerted to decrease the radiation dose to the population and to the environment. Thus, in making abatement and remediation policy decisions, not only economic costs but also human and environmental risk assessments are desired. This paper introduces a general framework for risk assessment and remedial policy evaluation using predictive modeling. Ecological risk assessment requires evaluation of the radionuclide distribution in ecosystems. The FORESTPATH model is used for predicting the radionuclide fate in forest compartments after deposition as well as for evaluating the efficiency of remedial policies. Time of intervention and radionuclide deposition profile was predicted as being crucial for the remediation efficiency. Risk assessment conducted for a critical group of forest users in Belarus shows that consumption of forest products (berries and mushrooms) leads to about 0.004% risk of a fatal cancer annually. Cost-benefit analysis for forest cleanup suggests that complete removal of organic layer is too expensive for application in Belarus and a better methodology is required. In conclusion, FORESTPATH modeling framework could have wide applications in environmental remediation of radionuclides and toxic metals as well as in dose reconstruction and, risk-assessment

Incorporating uncertainty in predictive species distribution modelling.

Science.gov (United States)

Beale, Colin M; Lennon, Jack J

2012-01-19

Motivated by the need to solve ecological problems (climate change, habitat fragmentation and biological invasions), there has been increasing interest in species distribution models (SDMs). Predictions from these models inform conservation policy, invasive species management and disease-control measures. However, predictions are subject to uncertainty, the degree and source of which is often unrecognized. Here, we review the SDM literature in the context of uncertainty, focusing on three main classes of SDM: niche-based models, demographic models and process-based models. We identify sources of uncertainty for each class and discuss how uncertainty can be minimized or included in the modelling process to give realistic measures of confidence around predictions. Because this has typically not been performed, we conclude that uncertainty in SDMs has often been underestimated and a false precision assigned to predictions of geographical distribution. We identify areas where development of new statistical tools will improve predictions from distribution models, notably the development of hierarchical models that link different types of distribution model and their attendant uncertainties across spatial scales. Finally, we discuss the need to develop more defensible methods for assessing predictive performance, quantifying model goodness-of-fit and for assessing the significance of model covariates.

Impact of implementation choices on quantitative predictions of cell-based computational models

Science.gov (United States)

Kursawe, Jochen; Baker, Ruth E.; Fletcher, Alexander G.

2017-09-01

'Cell-based' models provide a powerful computational tool for studying the mechanisms underlying the growth and dynamics of biological tissues in health and disease. An increasing amount of quantitative data with cellular resolution has paved the way for the quantitative parameterisation and validation of such models. However, the numerical implementation of cell-based models remains challenging, and little work has been done to understand to what extent implementation choices may influence model predictions. Here, we consider the numerical implementation of a popular class of cell-based models called vertex models, which are often used to study epithelial tissues. In two-dimensional vertex models, a tissue is approximated as a tessellation of polygons and the vertices of these polygons move due to mechanical forces originating from the cells. Such models have been used extensively to study the mechanical regulation of tissue topology in the literature. Here, we analyse how the model predictions may be affected by numerical parameters, such as the size of the time step, and non-physical model parameters, such as length thresholds for cell rearrangement. We find that vertex positions and summary statistics are sensitive to several of these implementation parameters. For example, the predicted tissue size decreases with decreasing cell cycle durations, and cell rearrangement may be suppressed by large time steps. These findings are counter-intuitive and illustrate that model predictions need to be thoroughly analysed and implementation details carefully considered when applying cell-based computational models in a quantitative setting.

EFFECT OF MEASUREMENT ERRORS ON PREDICTED COSMOLOGICAL CONSTRAINTS FROM SHEAR PEAK STATISTICS WITH LARGE SYNOPTIC SURVEY TELESCOPE

Energy Technology Data Exchange (ETDEWEB)

Bard, D.; Chang, C.; Kahn, S. M.; Gilmore, K.; Marshall, S. [KIPAC, Stanford University, 452 Lomita Mall, Stanford, CA 94309 (United States); Kratochvil, J. M.; Huffenberger, K. M. [Department of Physics, University of Miami, Coral Gables, FL 33124 (United States); May, M. [Physics Department, Brookhaven National Laboratory, Upton, NY 11973 (United States); AlSayyad, Y.; Connolly, A.; Gibson, R. R.; Jones, L.; Krughoff, S. [Department of Astronomy, University of Washington, Seattle, WA 98195 (United States); Ahmad, Z.; Bankert, J.; Grace, E.; Hannel, M.; Lorenz, S. [Department of Physics, Purdue University, West Lafayette, IN 47907 (United States); Haiman, Z.; Jernigan, J. G., E-mail: djbard@slac.stanford.edu [Department of Astronomy and Astrophysics, Columbia University, New York, NY 10027 (United States); and others

2013-09-01

We study the effect of galaxy shape measurement errors on predicted cosmological constraints from the statistics of shear peak counts with the Large Synoptic Survey Telescope (LSST). We use the LSST Image Simulator in combination with cosmological N-body simulations to model realistic shear maps for different cosmological models. We include both galaxy shape noise and, for the first time, measurement errors on galaxy shapes. We find that the measurement errors considered have relatively little impact on the constraining power of shear peak counts for LSST.

Models of alien species richness show moderate predictive accuracy and poor transferability

Directory of Open Access Journals (Sweden)

César Capinha

2018-06-01

Full Text Available Robust predictions of alien species richness are useful to assess global biodiversity change. Nevertheless, the capacity to predict spatial patterns of alien species richness remains largely unassessed. Using 22 data sets of alien species richness from diverse taxonomic groups and covering various parts of the world, we evaluated whether different statistical models were able to provide useful predictions of absolute and relative alien species richness, as a function of explanatory variables representing geographical, environmental and socio-economic factors. Five state-of-the-art count data modelling techniques were used and compared: Poisson and negative binomial generalised linear models (GLMs, multivariate adaptive regression splines (MARS, random forests (RF and boosted regression trees (BRT. We found that predictions of absolute alien species richness had a low to moderate accuracy in the region where the models were developed and a consistently poor accuracy in new regions. Predictions of relative richness performed in a superior manner in both geographical settings, but still were not good. Flexible tree ensembles-type techniques (RF and BRT were shown to be significantly better in modelling alien species richness than parametric linear models (such as GLM, despite the latter being more commonly applied for this purpose. Importantly, the poor spatial transferability of models also warrants caution in assuming the generality of the relationships they identify, e.g. by applying projections under future scenario conditions. Ultimately, our results strongly suggest that predictability of spatial variation in richness of alien species richness is limited. The somewhat more robust ability to rank regions according to the number of aliens they have (i.e. relative richness, suggests that models of aliens species richness may be useful for prioritising and comparing regions, but not for predicting exact species numbers.

Predictive user modeling with actionable attributes

NARCIS (Netherlands)

Zliobaite, I.; Pechenizkiy, M.

2013-01-01

Different machine learning techniques have been proposed and used for modeling individual and group user needs, interests and preferences. In the traditional predictive modeling instances are described by observable variables, called attributes. The goal is to learn a model for predicting the target

Prediction and assimilation of surf-zone processes using a Bayesian network: Part I: Forward models

Science.gov (United States)

Plant, Nathaniel G.; Holland, K. Todd

2011-01-01

Prediction of coastal processes, including waves, currents, and sediment transport, can be obtained from a variety of detailed geophysical-process models with many simulations showing significant skill. This capability supports a wide range of research and applied efforts that can benefit from accurate numerical predictions. However, the predictions are only as accurate as the data used to drive the models and, given the large temporal and spatial variability of the surf zone, inaccuracies in data are unavoidable such that useful predictions require corresponding estimates of uncertainty. We demonstrate how a Bayesian-network model can be used to provide accurate predictions of wave-height evolution in the surf zone given very sparse and/or inaccurate boundary-condition data. The approach is based on a formal treatment of a data-assimilation problem that takes advantage of significant reduction of the dimensionality of the model system. We demonstrate that predictions of a detailed geophysical model of the wave evolution are reproduced accurately using a Bayesian approach. In this surf-zone application, forward prediction skill was 83%, and uncertainties in the model inputs were accurately transferred to uncertainty in output variables. We also demonstrate that if modeling uncertainties were not conveyed to the Bayesian network (i.e., perfect data or model were assumed), then overly optimistic prediction uncertainties were computed. More consistent predictions and uncertainties were obtained by including model-parameter errors as a source of input uncertainty. Improved predictions (skill of 90%) were achieved because the Bayesian network simultaneously estimated optimal parameters while predicting wave heights.

Numerical flow simulation and efficiency prediction for axial turbines by advanced turbulence models

International Nuclear Information System (INIS)

Jošt, D; Škerlavaj, A; Lipej, A

2012-01-01

Numerical prediction of an efficiency of a 6-blade Kaplan turbine is presented. At first, the results of steady state analysis performed by different turbulence models for different operating regimes are compared to the measurements. For small and optimal angles of runner blades the efficiency was quite accurately predicted, but for maximal blade angle the discrepancy between calculated and measured values was quite large. By transient analysis, especially when the Scale Adaptive Simulation Shear Stress Transport (SAS SST) model with zonal Large Eddy Simulation (ZLES) in the draft tube was used, the efficiency was significantly improved. The improvement was at all operating points, but it was the largest for maximal discharge. The reason was better flow simulation in the draft tube. Details about turbulent structure in the draft tube obtained by SST, SAS SST and SAS SST with ZLES are illustrated in order to explain the reasons for differences in flow energy losses obtained by different turbulence models.

Numerical flow simulation and efficiency prediction for axial turbines by advanced turbulence models

Science.gov (United States)

Jošt, D.; Škerlavaj, A.; Lipej, A.

2012-11-01

Numerical prediction of an efficiency of a 6-blade Kaplan turbine is presented. At first, the results of steady state analysis performed by different turbulence models for different operating regimes are compared to the measurements. For small and optimal angles of runner blades the efficiency was quite accurately predicted, but for maximal blade angle the discrepancy between calculated and measured values was quite large. By transient analysis, especially when the Scale Adaptive Simulation Shear Stress Transport (SAS SST) model with zonal Large Eddy Simulation (ZLES) in the draft tube was used, the efficiency was significantly improved. The improvement was at all operating points, but it was the largest for maximal discharge. The reason was better flow simulation in the draft tube. Details about turbulent structure in the draft tube obtained by SST, SAS SST and SAS SST with ZLES are illustrated in order to explain the reasons for differences in flow energy losses obtained by different turbulence models.

Are more complex physiological models of forest ecosystems better choices for plot and regional predictions?

Science.gov (United States)

Wenchi Jin; Hong S. He; Frank R. Thompson

2016-01-01

Process-based forest ecosystem models vary from simple physiological, complex physiological, to hybrid empirical-physiological models. Previous studies indicate that complex models provide the best prediction at plot scale with a temporal extent of less than 10 years, however, it is largely untested as to whether complex models outperform the other two types of models...

A model for steady-state large-volume plasma generation

International Nuclear Information System (INIS)

Uhm, H.S.; Miller, J.D.; Schneider, R.F.

1991-01-01

In this paper, a simple, new scheme to generate a uniform, steady-state, large-volume plasma is presented. The weakly magnetized plasma is created by direct ionization of the background gas by low-energy electrons generated from thermionic filaments. An annular arrangement of the filaments ensures a uniform plasma density in the radial direction as predicted by theory. Experiments have been performed to characterize the plasma generated in such a configuration. In order to explain the experimental observation, we develop a bulk plasma theory based on plasma transport via cross-field diffusion. As assumed in the theoretical model, the experimental measurements indicate a uniform plasma density along the axis. Both the theory and experiment indicate that the plasma density is a function of the square of the external magnetic field. The theory also predicts the plasma density to be proportional to the neutral density to the two-thirds power in agreement with the experimental data. We also observe the experimental data to agree remarkably well with theoretical prediction for a broad range of system parameters

Predicting human chronically paralyzed muscle force: a comparison of three mathematical models.

Science.gov (United States)

Frey Law, Laura A; Shields, Richard K

2006-03-01

Chronic spinal cord injury (SCI) induces detrimental musculoskeletal adaptations that adversely affect health status, ranging from muscle paralysis and skin ulcerations to osteoporosis. SCI rehabilitative efforts may increasingly focus on preserving the integrity of paralyzed extremities to maximize health quality using electrical stimulation for isometric training and/or functional activities. Subject-specific mathematical muscle models could prove valuable for predicting the forces necessary to achieve therapeutic loading conditions in individuals with paralyzed limbs. Although numerous muscle models are available, three modeling approaches were chosen that can accommodate a variety of stimulation input patterns. To our knowledge, no direct comparisons between models using paralyzed muscle have been reported. The three models include 1) a simple second-order linear model with three parameters and 2) two six-parameter nonlinear models (a second-order nonlinear model and a Hill-derived nonlinear model). Soleus muscle forces from four individuals with complete, chronic SCI were used to optimize each model's parameters (using an increasing and decreasing frequency ramp) and to assess the models' predictive accuracies for constant and variable (doublet) stimulation trains at 5, 10, and 20 Hz in each individual. Despite the large differences in modeling approaches, the mean predicted force errors differed only moderately (8-15% error; P=0.0042), suggesting physiological force can be adequately represented by multiple mathematical constructs. The two nonlinear models predicted specific force characteristics better than the linear model in nearly all stimulation conditions, with minimal differences between the two nonlinear models. Either nonlinear mathematical model can provide reasonable force estimates; individual application needs may dictate the preferred modeling strategy.

Large animal models for vaccine development and testing.

Science.gov (United States)

Gerdts, Volker; Wilson, Heather L; Meurens, Francois; van Drunen Littel-van den Hurk, Sylvia; Wilson, Don; Walker, Stewart; Wheler, Colette; Townsend, Hugh; Potter, Andrew A

2015-01-01

The development of human vaccines continues to rely on the use of animals for research. Regulatory authorities require novel vaccine candidates to undergo preclinical assessment in animal models before being permitted to enter the clinical phase in human subjects. Substantial progress has been made in recent years in reducing and replacing the number of animals used for preclinical vaccine research through the use of bioinformatics and computational biology to design new vaccine candidates. However, the ultimate goal of a new vaccine is to instruct the immune system to elicit an effective immune response against the pathogen of interest, and no alternatives to live animal use currently exist for evaluation of this response. Studies identifying the mechanisms of immune protection; determining the optimal route and formulation of vaccines; establishing the duration and onset of immunity, as well as the safety and efficacy of new vaccines, must be performed in a living system. Importantly, no single animal model provides all the information required for advancing a new vaccine through the preclinical stage, and research over the last two decades has highlighted that large animals more accurately predict vaccine outcome in humans than do other models. Here we review the advantages and disadvantages of large animal models for human vaccine development and demonstrate that much of the success in bringing a new vaccine to market depends on choosing the most appropriate animal model for preclinical testing. © The Author 2015. Published by Oxford University Press on behalf of the Institute for Laboratory Animal Research. All rights reserved. For permissions, please email: journals.permissions@oup.com.

Improving Wind Farm Dispatchability Using Model Predictive Control for Optimal Operation of Grid-Scale Energy Storage

Directory of Open Access Journals (Sweden)

Douglas Halamay

2014-09-01

Full Text Available This paper demonstrates the use of model-based predictive control for energy storage systems to improve the dispatchability of wind power plants. Large-scale wind penetration increases the variability of power flow on the grid, thus increasing reserve requirements. Large energy storage systems collocated with wind farms can improve dispatchability of the wind plant by storing energy during generation over-the-schedule and sourcing energy during generation under-the-schedule, essentially providing on-site reserves. Model predictive control (MPC provides a natural framework for this application. By utilizing an accurate energy storage system model, control actions can be planned in the context of system power and state-of-charge limitations. MPC also enables the inclusion of predicted wind farm performance over a near-term horizon that allows control actions to be planned in anticipation of fast changes, such as wind ramps. This paper demonstrates that model-based predictive control can improve system performance compared with a standard non-predictive, non-model-based control approach. It is also demonstrated that secondary objectives, such as reducing the rate of change of the wind plant output (i.e., ramps, can be considered and successfully implemented within the MPC framework. Specifically, it is shown that scheduling error can be reduced by 81%, reserve requirements can be improved by up to 37%, and the number of ramp events can be reduced by 74%.

Decentralized Model Predictive Control for Cooperative Multiple Vehicles Subject to Communication Loss

Directory of Open Access Journals (Sweden)

Hojjat A. Izadi

2011-01-01

Full Text Available The decentralized model predictive control (DMPC of multiple cooperative vehicles with the possibility of communication loss/delay is investigated. The neighboring vehicles exchange their predicted trajectories at every sample time to maintain the cooperation objectives. In the event of a communication loss (packet dropout, the most recent available information, which is potentially delayed, is used. Then the communication loss problem changes to a cooperative problem when random large communication delays are present. Such large communication delays can lead to poor cooperation performance and unsafe behaviors such as collisions. A new DMPC approach is developed to improve the cooperation performance and achieve safety in the presence of the large communication delays. The proposed DMPC architecture estimates the tail of neighbor's trajectory which is not available due to the large communication delays for improving the performance. The concept of the tube MPC is also employed to provide the safety of the fleet against collisions, in the presence of large intervehicle communication delays. In this approach, a tube shaped trajectory set is assumed around the trajectory of the neighboring vehicles whose trajectory is delayed/lost. The radius of tube is a function of the communication delay and vehicle's maneuverability (in the absence of model uncertainty. The simulation of formation problem of multiple vehicles is employed to illustrate the effectiveness of the proposed approach.

Large-scale structural and textual similarity-based mining of knowledge graph to predict drug-drug interactions

KAUST Repository

Abdelaziz, Ibrahim; Fokoue, Achille; Hassanzadeh, Oktie; Zhang, Ping; Sadoghi, Mohammad

2017-01-01

Drug-Drug Interactions (DDIs) are a major cause of preventable Adverse Drug Reactions (ADRs), causing a significant burden on the patients’ health and the healthcare system. It is widely known that clinical studies cannot sufficiently and accurately identify DDIs for new drugs before they are made available on the market. In addition, existing public and proprietary sources of DDI information are known to be incomplete and/or inaccurate and so not reliable. As a result, there is an emerging body of research on in-silico prediction of drug-drug interactions. In this paper, we present Tiresias, a large-scale similarity-based framework that predicts DDIs through link prediction. Tiresias takes in various sources of drug-related data and knowledge as inputs, and provides DDI predictions as outputs. The process starts with semantic integration of the input data that results in a knowledge graph describing drug attributes and relationships with various related entities such as enzymes, chemical structures, and pathways. The knowledge graph is then used to compute several similarity measures between all the drugs in a scalable and distributed framework. In particular, Tiresias utilizes two classes of features in a knowledge graph: local and global features. Local features are derived from the information directly associated to each drug (i.e., one hop away) while global features are learnt by minimizing a global loss function that considers the complete structure of the knowledge graph. The resulting similarity metrics are used to build features for a large-scale logistic regression model to predict potential DDIs. We highlight the novelty of our proposed Tiresias and perform thorough evaluation of the quality of the predictions. The results show the effectiveness of Tiresias in both predicting new interactions among existing drugs as well as newly developed drugs.

Large-scale structural and textual similarity-based mining of knowledge graph to predict drug-drug interactions

KAUST Repository

Abdelaziz, Ibrahim

2017-06-12

Drug-Drug Interactions (DDIs) are a major cause of preventable Adverse Drug Reactions (ADRs), causing a significant burden on the patients’ health and the healthcare system. It is widely known that clinical studies cannot sufficiently and accurately identify DDIs for new drugs before they are made available on the market. In addition, existing public and proprietary sources of DDI information are known to be incomplete and/or inaccurate and so not reliable. As a result, there is an emerging body of research on in-silico prediction of drug-drug interactions. In this paper, we present Tiresias, a large-scale similarity-based framework that predicts DDIs through link prediction. Tiresias takes in various sources of drug-related data and knowledge as inputs, and provides DDI predictions as outputs. The process starts with semantic integration of the input data that results in a knowledge graph describing drug attributes and relationships with various related entities such as enzymes, chemical structures, and pathways. The knowledge graph is then used to compute several similarity measures between all the drugs in a scalable and distributed framework. In particular, Tiresias utilizes two classes of features in a knowledge graph: local and global features. Local features are derived from the information directly associated to each drug (i.e., one hop away) while global features are learnt by minimizing a global loss function that considers the complete structure of the knowledge graph. The resulting similarity metrics are used to build features for a large-scale logistic regression model to predict potential DDIs. We highlight the novelty of our proposed Tiresias and perform thorough evaluation of the quality of the predictions. The results show the effectiveness of Tiresias in both predicting new interactions among existing drugs as well as newly developed drugs.

A Method to Quantify Plant Availability and Initiating Event Frequency Using a Large Event Tree, Small Fault Tree Model

International Nuclear Information System (INIS)

Kee, Ernest J.; Sun, Alice; Rodgers, Shawn; Popova, ElmiraV; Nelson, Paul; Moiseytseva, Vera; Wang, Eric

2006-01-01

South Texas Project uses a large fault tree to produce scenarios (minimal cut sets) used in quantification of plant availability and event frequency predictions. On the other hand, the South Texas Project probabilistic risk assessment model uses a large event tree, small fault tree for quantifying core damage and radioactive release frequency predictions. The South Texas Project is converting its availability and event frequency model to use a large event tree, small fault in an effort to streamline application support and to provide additional detail in results. The availability and event frequency model as well as the applications it supports (maintenance and operational risk management, system engineering health assessment, preventive maintenance optimization, and RIAM) are briefly described. A methodology to perform availability modeling in a large event tree, small fault tree framework is described in detail. How the methodology can be used to support South Texas Project maintenance and operations risk management is described in detail. Differences with other fault tree methods and other recently proposed methods are discussed in detail. While the methods described are novel to the South Texas Project Risk Management program and to large event tree, small fault tree models, concepts in the area of application support and availability modeling have wider applicability to the industry. (authors)

Prediction of stably stratified homogeneous shear flows with second-order turbulence models

International Nuclear Information System (INIS)

Pereira, J C F; Rocha, J M P

2010-01-01

The present study investigated the role of pressure-correlation second-order turbulence modelling schemes on the predicted behaviour of stably stratified homogeneous vertical-sheared turbulence. The pressure-correlation terms were modelled with a nonlinear formulation (Craft 1991), which was compared with a linear pressure-strain model and the 'isotropization of production' model for the pressure-scalar correlation. Two additional modelling issues were investigated: the influence of the buoyancy term in the kinetic energy dissipation rate equation and the time scale in the thermal production term in the scalar variance dissipation equation. The predicted effects of increasing the Richardson number on turbulence characteristics were compared against a comprehensive set of direct numerical simulation databases. The linear models provide a broadly satisfactory description of the major effects of the Richardson number on stratified shear flow. The buoyancy term in the dissipation equation of the turbulent kinetic energy generates excessively low levels of dissipation. For moderate and large Richardson numbers, the term yields unrealistic linear oscillations in the shear and buoyancy production terms, and therefore should be dropped in this flow (or at least their coefficient c ε3 should be substantially reduced from its standard value). The mechanical dissipation time scale provides marginal improvements in comparison to the scalar time scale in the production. The observed inaccuracy of the linear model in predicting the magnitude of the effects on the velocity anisotropy was demonstrated to be attributed mainly to the defective behaviour of the pressure-correlation model, especially for stronger stratification. The turbulence closure embodying a nonlinear formulation for the pressure-correlations and specific versions of the dissipation equations failed to predict the tendency of the flow to anisotropy with increasing stratification. By isolating the effects of the

MJO prediction skill of the subseasonal-to-seasonal (S2S) prediction models

Science.gov (United States)

Son, S. W.; Lim, Y.; Kim, D.

2017-12-01

The Madden-Julian Oscillation (MJO), the dominant mode of tropical intraseasonal variability, provides the primary source of tropical and extratropical predictability on subseasonal to seasonal timescales. To better understand its predictability, this study conducts quantitative evaluation of MJO prediction skill in the state-of-the-art operational models participating in the subseasonal-to-seasonal (S2S) prediction project. Based on bivariate correlation coefficient of 0.5, the S2S models exhibit MJO prediction skill ranging from 12 to 36 days. These prediction skills are affected by both the MJO amplitude and phase errors, the latter becoming more important with forecast lead times. Consistent with previous studies, the MJO events with stronger initial amplitude are typically better predicted. However, essentially no sensitivity to the initial MJO phase is observed. Overall MJO prediction skill and its inter-model spread are further related with the model mean biases in moisture fields and longwave cloud-radiation feedbacks. In most models, a dry bias quickly builds up in the deep tropics, especially across the Maritime Continent, weakening horizontal moisture gradient. This likely dampens the organization and propagation of MJO. Most S2S models also underestimate the longwave cloud-radiation feedbacks in the tropics, which may affect the maintenance of the MJO convective envelop. In general, the models with a smaller bias in horizontal moisture gradient and longwave cloud-radiation feedbacks show a higher MJO prediction skill, suggesting that improving those processes would enhance MJO prediction skill.

Evaluating Predictive Uncertainty of Hyporheic Exchange Modelling

Science.gov (United States)

Chow, R.; Bennett, J.; Dugge, J.; Wöhling, T.; Nowak, W.

2017-12-01

Hyporheic exchange is the interaction of water between rivers and groundwater, and is difficult to predict. One of the largest contributions to predictive uncertainty for hyporheic fluxes have been attributed to the representation of heterogeneous subsurface properties. This research aims to evaluate which aspect of the subsurface representation - the spatial distribution of hydrofacies or the model for local-scale (within-facies) heterogeneity - most influences the predictive uncertainty. Also, we seek to identify data types that help reduce this uncertainty best. For this investigation, we conduct a modelling study of the Steinlach River meander, in Southwest Germany. The Steinlach River meander is an experimental site established in 2010 to monitor hyporheic exchange at the meander scale. We use HydroGeoSphere, a fully integrated surface water-groundwater model, to model hyporheic exchange and to assess the predictive uncertainty of hyporheic exchange transit times (HETT). A highly parameterized complex model is built and treated as `virtual reality', which is in turn modelled with simpler subsurface parameterization schemes (Figure). Then, we conduct Monte-Carlo simulations with these models to estimate the predictive uncertainty. Results indicate that: Uncertainty in HETT is relatively small for early times and increases with transit times. Uncertainty from local-scale heterogeneity is negligible compared to uncertainty in the hydrofacies distribution. Introducing more data to a poor model structure may reduce predictive variance, but does not reduce predictive bias. Hydraulic head observations alone cannot constrain the uncertainty of HETT, however an estimate of hyporheic exchange flux proves to be more effective at reducing this uncertainty. Figure: Approach for evaluating predictive model uncertainty. A conceptual model is first developed from the field investigations. A complex model (`virtual reality') is then developed based on that conceptual model

A dynamic capacity degradation model and its applications considering varying load for a large format Li-ion battery

International Nuclear Information System (INIS)

Ouyang, Minggao; Feng, Xuning; Han, Xuebing; Lu, Languang; Li, Zhe; He, Xiangming

2016-01-01

Highlights: • A dynamic capacity degradation model for large format Li-ion battery is proposed. • The change of the model parameters directly link with the degradation mechanisms. • The model can simulate the fading behavior of Li-ion battery under varying loads. • The model can help evaluate the longevity of a battery system under specific load. • The model can help predict the evolution of cell variations within a battery pack. - Abstract: The capacity degradation of the lithium ion battery should be well predicted during battery system design. Therefore, high-fidelity capacity degradation models that are suitable for the task of capacity prediction are required. This paper proposes a novel capacity degradation model that can simulate the degradation dynamics under varying working conditions for large-format lithium ion batteries. The degradation model is built based on a mechanistic and prognostic model (MPM) whose parameters are closely linked with the degradation mechanisms of lithium ion batteries. Chemical kinetics was set to drive the parameters of the MPM to change as capacity degradation continues. With the dynamic parameters of the MPM, the capacity predicted by the degradation model decreases as the cycle continues. Accelerated aging tests were conducted on three types of commercial lithium ion batteries to calibrate the capacity degradation model. The good fit with the experimental data indicates that the model can capture the degradation mechanisms well for different types of commercial lithium ion batteries. Furthermore, the calibrated model can be used to (1) evaluate the longevity of a battery system under a specific working load and (2) predict the evolution of cell variations within a battery pack when different cell works at different conditions. Correlated applications are discussed using the calibrated degradation model.

An adaptive two-stage analog/regression model for probabilistic prediction of small-scale precipitation in France

Science.gov (United States)

Chardon, Jérémy; Hingray, Benoit; Favre, Anne-Catherine

2018-01-01

Statistical downscaling models (SDMs) are often used to produce local weather scenarios from large-scale atmospheric information. SDMs include transfer functions which are based on a statistical link identified from observations between local weather and a set of large-scale predictors. As physical processes driving surface weather vary in time, the most relevant predictors and the regression link are likely to vary in time too. This is well known for precipitation for instance and the link is thus often estimated after some seasonal stratification of the data. In this study, we present a two-stage analog/regression model where the regression link is estimated from atmospheric analogs of the current prediction day. Atmospheric analogs are identified from fields of geopotential heights at 1000 and 500 hPa. For the regression stage, two generalized linear models are further used to model the probability of precipitation occurrence and the distribution of non-zero precipitation amounts, respectively. The two-stage model is evaluated for the probabilistic prediction of small-scale precipitation over France. It noticeably improves the skill of the prediction for both precipitation occurrence and amount. As the analog days vary from one prediction day to another, the atmospheric predictors selected in the regression stage and the value of the corresponding regression coefficients can vary from one prediction day to another. The model allows thus for a day-to-day adaptive and tailored downscaling. It can also reveal specific predictors for peculiar and non-frequent weather configurations.

Improvement of PM10 prediction in East Asia using inverse modeling

Science.gov (United States)

Koo, Youn-Seo; Choi, Dae-Ryun; Kwon, Hi-Yong; Jang, Young-Kee; Han, Jin-Seok

2015-04-01

Aerosols from anthropogenic emissions in industrialized region in China as well as dust emissions from southern Mongolia and northern China that transport along prevailing northwestern wind have a large influence on the air quality in Korea. The emission inventory in the East Asia region is an important factor in chemical transport modeling (CTM) for PM10 (particulate matters less than 10 ㎛ in aerodynamic diameter) forecasts and air quality management in Korea. Most previous studies showed that predictions of PM10 mass concentration by the CTM were underestimated when comparing with observational data. In order to fill the gap in discrepancies between observations and CTM predictions, the inverse Bayesian approach with Comprehensive Air-quality Model with extension (CAMx) forward model was applied to obtain optimized a posteriori PM10 emissions in East Asia. The predicted PM10 concentrations with a priori emission were first compared with observations at monitoring sites in China and Korea for January and August 2008. The comparison showed that PM10 concentrations with a priori PM10 emissions for anthropogenic and dust sources were generally under-predicted. The result from the inverse modeling indicated that anthropogenic PM10 emissions in the industrialized and urbanized areas in China were underestimated while dust emissions from desert and barren soil in southern Mongolia and northern China were overestimated. A priori PM10 emissions from northeastern China regions including Shenyang, Changchun, and Harbin were underestimated by about 300% (i.e., the ratio of a posteriori to a priori PM10 emission was a factor of about 3). The predictions of PM10 concentrations with a posteriori emission showed better agreement with the observations, implying that the inverse modeling minimized the discrepancies in the model predictions by improving PM10 emissions in East Asia.

Modeling, robust and distributed model predictive control for freeway networks

NARCIS (Netherlands)

Liu, S.

2016-01-01

In Model Predictive Control (MPC) for traffic networks, traffic models are crucial since they are used as prediction models for determining the optimal control actions. In order to reduce the computational complexity of MPC for traffic networks, macroscopic traffic models are often used instead of

Novel patch modelling method for efficient simulation and prediction uncertainty analysis of multi-scale groundwater flow and transport processes

Science.gov (United States)

Sreekanth, J.; Moore, Catherine

2018-04-01

The application of global sensitivity and uncertainty analysis techniques to groundwater models of deep sedimentary basins are typically challenged by large computational burdens combined with associated numerical stability issues. The highly parameterized approaches required for exploring the predictive uncertainty associated with the heterogeneous hydraulic characteristics of multiple aquifers and aquitards in these sedimentary basins exacerbate these issues. A novel Patch Modelling Methodology is proposed for improving the computational feasibility of stochastic modelling analysis of large-scale and complex groundwater models. The method incorporates a nested groundwater modelling framework that enables efficient simulation of groundwater flow and transport across multiple spatial and temporal scales. The method also allows different processes to be simulated within different model scales. Existing nested model methodologies are extended by employing 'joining predictions' for extrapolating prediction-salient information from one model scale to the next. This establishes a feedback mechanism supporting the transfer of information from child models to parent models as well as parent models to child models in a computationally efficient manner. This feedback mechanism is simple and flexible and ensures that while the salient small scale features influencing larger scale prediction are transferred back to the larger scale, this does not require the live coupling of models. This method allows the modelling of multiple groundwater flow and transport processes using separate groundwater models that are built for the appropriate spatial and temporal scales, within a stochastic framework, while also removing the computational burden associated with live model coupling. The utility of the method is demonstrated by application to an actual large scale aquifer injection scheme in Australia.

Staying Power of Churn Prediction Models

NARCIS (Netherlands)

Risselada, Hans; Verhoef, Peter C.; Bijmolt, Tammo H. A.

In this paper, we study the staying power of various churn prediction models. Staying power is defined as the predictive performance of a model in a number of periods after the estimation period. We examine two methods, logit models and classification trees, both with and without applying a bagging

Water and salt balance modelling to predict the effects of land-use changes in forested catchments. 1. Small catchment water balance model

Science.gov (United States)

Sivapalan, Murugesu; Ruprecht, John K.; Viney, Neil R.

1996-03-01

A long-term water balance model has been developed to predict the hydrological effects of land-use change (especially forest clearing) in small experimental catchments in the south-west of Western Australia. This small catchment model has been used as the building block for the development of a large catchment-scale model, and has also formed the basis for a coupled water and salt balance model, developed to predict the changes in stream salinity resulting from land-use and climate change. The application of the coupled salt and water balance model to predict stream salinities in two small experimental catchments, and the application of the large catchment-scale model to predict changes in water yield in a medium-sized catchment that is being mined for bauxite, are presented in Parts 2 and 3, respectively, of this series of papers.The small catchment model has been designed as a simple, robust, conceptually based model of the basic daily water balance fluxes in forested catchments. The responses of the catchment to rainfall and pan evaporation are conceptualized in terms of three interdependent subsurface stores A, B and F. Store A depicts a near-stream perched aquifer system; B represents a deeper, permanent groundwater system; and F is an intermediate, unsaturated infiltration store. The responses of these stores are characterized by a set of constitutive relations which involves a number of conceptual parameters. These parameters are estimated by calibration by comparing observed and predicted runoff. The model has performed very well in simulations carried out on Salmon and Wights, two small experimental catchments in the Collie River basin in south-west Western Australia. The results from the application of the model to these small catchments are presented in this paper.

Finite element modelling for fatigue stress analysis of large suspension bridges

Science.gov (United States)

Chan, Tommy H. T.; Guo, L.; Li, Z. X.

2003-03-01

Fatigue is an important failure mode for large suspension bridges under traffic loadings. However, large suspension bridges have so many attributes that it is difficult to analyze their fatigue damage using experimental measurement methods. Numerical simulation is a feasible method of studying such fatigue damage. In British standards, the finite element method is recommended as a rigorous method for steel bridge fatigue analysis. This paper aims at developing a finite element (FE) model of a large suspension steel bridge for fatigue stress analysis. As a case study, a FE model of the Tsing Ma Bridge is presented. The verification of the model is carried out with the help of the measured bridge modal characteristics and the online data measured by the structural health monitoring system installed on the bridge. The results show that the constructed FE model is efficient for bridge dynamic analysis. Global structural analyses using the developed FE model are presented to determine the components of the nominal stress generated by railway loadings and some typical highway loadings. The critical locations in the bridge main span are also identified with the numerical results of the global FE stress analysis. Local stress analysis of a typical weld connection is carried out to obtain the hot-spot stresses in the region. These results provide a basis for evaluating fatigue damage and predicting the remaining life of the bridge.

Application of Fracture Distribution Prediction Model in Xihu Depression of East China Sea

Science.gov (United States)

Yan, Weifeng; Duan, Feifei; Zhang, Le; Li, Ming

2018-02-01

There are different responses on each of logging data with the changes of formation characteristics and outliers caused by the existence of fractures. For this reason, the development of fractures in formation can be characterized by the fine analysis of logging curves. The well logs such as resistivity, sonic transit time, density, neutron porosity and gamma ray, which are classified as conventional well logs, are more sensitive to formation fractures. In view of traditional fracture prediction model, using the simple weighted average of different logging data to calculate the comprehensive fracture index, are more susceptible to subjective factors and exist a large deviation, a statistical method is introduced accordingly. Combining with responses of conventional logging data on the development of formation fracture, a prediction model based on membership function is established, and its essence is to analyse logging data with fuzzy mathematics theory. The fracture prediction results in a well formation in NX block of Xihu depression through two models are compared with that of imaging logging, which shows that the accuracy of fracture prediction model based on membership function is better than that of traditional model. Furthermore, the prediction results are highly consistent with imaging logs and can reflect the development of cracks much better. It can provide a reference for engineering practice.

Bridge Structure Deformation Prediction Based on GNSS Data Using Kalman-ARIMA-GARCH Model.

Science.gov (United States)

Xin, Jingzhou; Zhou, Jianting; Yang, Simon X; Li, Xiaoqing; Wang, Yu

2018-01-19

Bridges are an essential part of the ground transportation system. Health monitoring is fundamentally important for the safety and service life of bridges. A large amount of structural information is obtained from various sensors using sensing technology, and the data processing has become a challenging issue. To improve the prediction accuracy of bridge structure deformation based on data mining and to accurately evaluate the time-varying characteristics of bridge structure performance evolution, this paper proposes a new method for bridge structure deformation prediction, which integrates the Kalman filter, autoregressive integrated moving average model (ARIMA), and generalized autoregressive conditional heteroskedasticity (GARCH). Firstly, the raw deformation data is directly pre-processed using the Kalman filter to reduce the noise. After that, the linear recursive ARIMA model is established to analyze and predict the structure deformation. Finally, the nonlinear recursive GARCH model is introduced to further improve the accuracy of the prediction. Simulation results based on measured sensor data from the Global Navigation Satellite System (GNSS) deformation monitoring system demonstrated that: (1) the Kalman filter is capable of denoising the bridge deformation monitoring data; (2) the prediction accuracy of the proposed Kalman-ARIMA-GARCH model is satisfactory, where the mean absolute error increases only from 3.402 mm to 5.847 mm with the increment of the prediction step; and (3) in comparision to the Kalman-ARIMA model, the Kalman-ARIMA-GARCH model results in superior prediction accuracy as it includes partial nonlinear characteristics (heteroscedasticity); the mean absolute error of five-step prediction using the proposed model is improved by 10.12%. This paper provides a new way for structural behavior prediction based on data processing, which can lay a foundation for the early warning of bridge health monitoring system based on sensor data using sensing

Bridge Structure Deformation Prediction Based on GNSS Data Using Kalman-ARIMA-GARCH Model

Directory of Open Access Journals (Sweden)

Jingzhou Xin

2018-01-01

Full Text Available Bridges are an essential part of the ground transportation system. Health monitoring is fundamentally important for the safety and service life of bridges. A large amount of structural information is obtained from various sensors using sensing technology, and the data processing has become a challenging issue. To improve the prediction accuracy of bridge structure deformation based on data mining and to accurately evaluate the time-varying characteristics of bridge structure performance evolution, this paper proposes a new method for bridge structure deformation prediction, which integrates the Kalman filter, autoregressive integrated moving average model (ARIMA, and generalized autoregressive conditional heteroskedasticity (GARCH. Firstly, the raw deformation data is directly pre-processed using the Kalman filter to reduce the noise. After that, the linear recursive ARIMA model is established to analyze and predict the structure deformation. Finally, the nonlinear recursive GARCH model is introduced to further improve the accuracy of the prediction. Simulation results based on measured sensor data from the Global Navigation Satellite System (GNSS deformation monitoring system demonstrated that: (1 the Kalman filter is capable of denoising the bridge deformation monitoring data; (2 the prediction accuracy of the proposed Kalman-ARIMA-GARCH model is satisfactory, where the mean absolute error increases only from 3.402 mm to 5.847 mm with the increment of the prediction step; and (3 in comparision to the Kalman-ARIMA model, the Kalman-ARIMA-GARCH model results in superior prediction accuracy as it includes partial nonlinear characteristics (heteroscedasticity; the mean absolute error of five-step prediction using the proposed model is improved by 10.12%. This paper provides a new way for structural behavior prediction based on data processing, which can lay a foundation for the early warning of bridge health monitoring system based on sensor data

Large-scale assessment of the zebrafish embryo as a possible predictive model in toxicity testing.

Directory of Open Access Journals (Sweden)

Shaukat Ali

Full Text Available BACKGROUND: In the drug discovery pipeline, safety pharmacology is a major issue. The zebrafish has been proposed as a model that can bridge the gap in this field between cell assays (which are cost-effective, but low in data content and rodent assays (which are high in data content, but less cost-efficient. However, zebrafish assays are only likely to be useful if they can be shown to have high predictive power. We examined this issue by assaying 60 water-soluble compounds representing a range of chemical classes and toxicological mechanisms. METHODOLOGY/PRINCIPAL FINDINGS: Over 20,000 wild-type zebrafish embryos (including controls were cultured individually in defined buffer in 96-well plates. Embryos were exposed for a 96 hour period starting at 24 hours post fertilization. A logarithmic concentration series was used for range-finding, followed by a narrower geometric series for LC(50 determination. Zebrafish embryo LC(50 (log mmol/L, and published data on rodent LD(50 (log mmol/kg, were found to be strongly correlated (using Kendall's rank correlation tau and Pearson's product-moment correlation. The slope of the regression line for the full set of compounds was 0.73403. However, we found that the slope was strongly influenced by compound class. Thus, while most compounds had a similar toxicity level in both species, some compounds were markedly more toxic in zebrafish than in rodents, or vice versa. CONCLUSIONS: For the substances examined here, in aggregate, the zebrafish embryo model has good predictivity for toxicity in rodents. However, the correlation between zebrafish and rodent toxicity varies considerably between individual compounds and compound class. We discuss the strengths and limitations of the zebrafish model in light of these findings.

A dynamic programming approach for quickly estimating large network-based MEV models

DEFF Research Database (Denmark)

Mai, Tien; Frejinger, Emma; Fosgerau, Mogens

2017-01-01

We propose a way to estimate a family of static Multivariate Extreme Value (MEV) models with large choice sets in short computational time. The resulting model is also straightforward and fast to use for prediction. Following Daly and Bierlaire (2006), the correlation structure is defined by a ro...... to converge (4.3 h on an Intel(R) 3.2 GHz machine using a non-parallelized code). We also show that our approach allows to estimate a cross-nested logit model of 111 nests with a real data set of more than 100,000 observations in 14 h....

Economic decision making and the application of nonparametric prediction models

Science.gov (United States)

Attanasi, E.D.; Coburn, T.C.; Freeman, P.A.

2008-01-01

Sustained increases in energy prices have focused attention on gas resources in low-permeability shale or in coals that were previously considered economically marginal. Daily well deliverability is often relatively small, although the estimates of the total volumes of recoverable resources in these settings are often large. Planning and development decisions for extraction of such resources must be areawide because profitable extraction requires optimization of scale economies to minimize costs and reduce risk. For an individual firm, the decision to enter such plays depends on reconnaissance-level estimates of regional recoverable resources and on cost estimates to develop untested areas. This paper shows how simple nonparametric local regression models, used to predict technically recoverable resources at untested sites, can be combined with economic models to compute regional-scale cost functions. The context of the worked example is the Devonian Antrim-shale gas play in the Michigan basin. One finding relates to selection of the resource prediction model to be used with economic models. Models chosen because they can best predict aggregate volume over larger areas (many hundreds of sites) smooth out granularity in the distribution of predicted volumes at individual sites. This loss of detail affects the representation of economic cost functions and may affect economic decisions. Second, because some analysts consider unconventional resources to be ubiquitous, the selection and order of specific drilling sites may, in practice, be determined arbitrarily by extraneous factors. The analysis shows a 15-20% gain in gas volume when these simple models are applied to order drilling prospects strategically rather than to choose drilling locations randomly. Copyright ?? 2008 Society of Petroleum Engineers.

Aespoe Pillar Stability Experiment. Summary of preparatory work and predictive modelling

International Nuclear Information System (INIS)

Andersson, J. Christer

2004-11-01

The Aespoe Pillar Stability Experiment, APSE, is a large scale rock mechanics experiment for research of the spalling process and the possibility for numerical modelling of it. The experiment can be summarized in three objectives: Demonstrate the current capability to predict spalling in a fractured rock mass; Demonstrate the effect of backfill (confining pressure) on the rock mass response; and Comparison of 2D and 3D mechanical and thermal predicting capabilities. This report is a summary of the works that has been performed in the experiment prior to the heating of the rock mass. The major activities that have been performed and are discussed herein are: 1) The geology of the experiment drift in general and the experiment volume in particular. 2) The design process of the experiment and thoughts behind some of the important decisions. 3) The monitoring programme and the supporting constructions for the instruments. 4) The numerical modelling, approaches taken and a summary of the predictions. In the end of the report there is a comparison of the results from the different models. Included is also a comparison of the time needed for building, realizing and make changes in the different models

Life Prediction of Large Lithium-Ion Battery Packs with Active and Passive Balancing

Energy Technology Data Exchange (ETDEWEB)

Shi, Ying [National Renewable Energy Laboratory (NREL), Golden, CO (United States); Smith, Kandler A [National Renewable Energy Laboratory (NREL), Golden, CO (United States); Zane, Regan [Utah State University; Anderson, Dyche [Ford Motor Company

2017-07-03

Lithium-ion battery packs take a major part of large-scale stationary energy storage systems. One challenge in reducing battery pack cost is to reduce pack size without compromising pack service performance and lifespan. Prognostic life model can be a powerful tool to handle the state of health (SOH) estimate and enable active life balancing strategy to reduce cell imbalance and extend pack life. This work proposed a life model using both empirical and physical-based approaches. The life model described the compounding effect of different degradations on the entire cell with an empirical model. Then its lower-level submodels considered the complex physical links between testing statistics (state of charge level, C-rate level, duty cycles, etc.) and the degradation reaction rates with respect to specific aging mechanisms. The hybrid approach made the life model generic, robust and stable regardless of battery chemistry and application usage. The model was validated with a custom pack with both passive and active balancing systems implemented, which created four different aging paths in the pack. The life model successfully captured the aging trajectories of all four paths. The life model prediction errors on capacity fade and resistance growth were within +/-3% and +/-5% of the experiment measurements.

The Application Law of Large Numbers That Predicts The Amount of Actual Loss in Insurance of Life

Science.gov (United States)

Tinungki, Georgina Maria

2018-03-01

The law of large numbers is a statistical concept that calculates the average number of events or risks in a sample or population to predict something. The larger the population is calculated, the more accurate predictions. In the field of insurance, the Law of Large Numbers is used to predict the risk of loss or claims of some participants so that the premium can be calculated appropriately. For example there is an average that of every 100 insurance participants, there is one participant who filed an accident claim, then the premium of 100 participants should be able to provide Sum Assured to at least 1 accident claim. The larger the insurance participant is calculated, the more precise the prediction of the calendar and the calculation of the premium. Life insurance, as a tool for risk spread, can only work if a life insurance company is able to bear the same risk in large numbers. Here apply what is called the law of large number. The law of large numbers states that if the amount of exposure to losses increases, then the predicted loss will be closer to the actual loss. The use of the law of large numbers allows the number of losses to be predicted better.

An analysis of seasonal predictability in coupled model forecasts

Energy Technology Data Exchange (ETDEWEB)

Peng, P.; Wang, W. [NOAA, Climate Prediction Center, Washington, DC (United States); Kumar, A. [NOAA, Climate Prediction Center, Washington, DC (United States); NCEP/NWS/NOAA, Climate Prediction Center, Camp Springs, MD (United States)

2011-02-15

In the recent decade, operational seasonal prediction systems based on initialized coupled models have been developed. An analysis of how the predictability of seasonal means in the initialized coupled predictions evolves with lead-time is presented. Because of the short lead-time, such an analysis for the temporal behavior of seasonal predictability involves a mix of both the predictability of the first and the second kind. The analysis focuses on the lead-time dependence of ensemble mean variance, and the forecast spread. Further, the analysis is for a fixed target season of December-January-February, and is for sea surface temperature, rainfall, and 200-mb height. The analysis is based on a large set of hindcasts from an initialized coupled seasonal prediction system. Various aspects of predictability of the first and the second kind are highlighted for variables with long (for example, SST), and fast (for example, atmospheric) adjustment time scale. An additional focus of the analysis is how the predictability in the initialized coupled seasonal predictions compares with estimates based on the AMIP simulations. The results indicate that differences in the set up of AMIP simulations and coupled predictions, for example, representation of air-sea interactions, and evolution of forecast spread from initial conditions do not change fundamental conclusion about the seasonal predictability. A discussion of the analysis presented herein, and its implications for the use of AMIP simulations for climate attribution, and for time-slice experiments to provide regional information, is also included. (orig.)

Performance Modeling of Hybrid MPI/OpenMP Scientific Applications on Large-scale Multicore Cluster Systems

KAUST Repository

Wu, Xingfu; Taylor, Valerie

2011-01-01

In this paper, we present a performance modeling framework based on memory bandwidth contention time and a parameterized communication model to predict the performance of OpenMP, MPI and hybrid applications with weak scaling on three large-scale multicore clusters: IBM POWER4, POWER5+ and Blue Gene/P, and analyze the performance of these MPI, OpenMP and hybrid applications. We use STREAM memory benchmarks to provide initial performance analysis and model validation of MPI and OpenMP applications on these multicore clusters because the measured sustained memory bandwidth can provide insight into the memory bandwidth that a system should sustain on scientific applications with the same amount of workload per core. In addition to using these benchmarks, we also use a weak-scaling hybrid MPI/OpenMP large-scale scientific application: Gyro kinetic Toroidal Code in magnetic fusion to validate our performance model of the hybrid application on these multicore clusters. The validation results for our performance modeling method show less than 7.77% error rate in predicting the performance of hybrid MPI/OpenMP GTC on up to 512 cores on these multicore clusters. © 2011 IEEE.

Performance Modeling of Hybrid MPI/OpenMP Scientific Applications on Large-scale Multicore Cluster Systems

KAUST Repository

Wu, Xingfu

2011-08-01

In this paper, we present a performance modeling framework based on memory bandwidth contention time and a parameterized communication model to predict the performance of OpenMP, MPI and hybrid applications with weak scaling on three large-scale multicore clusters: IBM POWER4, POWER5+ and Blue Gene/P, and analyze the performance of these MPI, OpenMP and hybrid applications. We use STREAM memory benchmarks to provide initial performance analysis and model validation of MPI and OpenMP applications on these multicore clusters because the measured sustained memory bandwidth can provide insight into the memory bandwidth that a system should sustain on scientific applications with the same amount of workload per core. In addition to using these benchmarks, we also use a weak-scaling hybrid MPI/OpenMP large-scale scientific application: Gyro kinetic Toroidal Code in magnetic fusion to validate our performance model of the hybrid application on these multicore clusters. The validation results for our performance modeling method show less than 7.77% error rate in predicting the performance of hybrid MPI/OpenMP GTC on up to 512 cores on these multicore clusters. © 2011 IEEE.

A Bayesian Hierarchical Modeling Approach to Predicting Flow in Ungauged Basins

Science.gov (United States)

Gronewold, A.; Alameddine, I.; Anderson, R. M.

2009-12-01

Recent innovative approaches to identifying and applying regression-based relationships between land use patterns (such as increasing impervious surface area and decreasing vegetative cover) and rainfall-runoff model parameters represent novel and promising improvements to predicting flow from ungauged basins. In particular, these approaches allow for predicting flows under uncertain and potentially variable future conditions due to rapid land cover changes, variable climate conditions, and other factors. Despite the broad range of literature on estimating rainfall-runoff model parameters, however, the absence of a robust set of modeling tools for identifying and quantifying uncertainties in (and correlation between) rainfall-runoff model parameters represents a significant gap in current hydrological modeling research. Here, we build upon a series of recent publications promoting novel Bayesian and probabilistic modeling strategies for quantifying rainfall-runoff model parameter estimation uncertainty. Our approach applies alternative measures of rainfall-runoff model parameter joint likelihood (including Nash-Sutcliffe efficiency, among others) to simulate samples from the joint parameter posterior probability density function. We then use these correlated samples as response variables in a Bayesian hierarchical model with land use coverage data as predictor variables in order to develop a robust land use-based tool for forecasting flow in ungauged basins while accounting for, and explicitly acknowledging, parameter estimation uncertainty. We apply this modeling strategy to low-relief coastal watersheds of Eastern North Carolina, an area representative of coastal resource waters throughout the world because of its sensitive embayments and because of the abundant (but currently threatened) natural resources it hosts. Consequently, this area is the subject of several ongoing studies and large-scale planning initiatives, including those conducted through the United

Accuracy assessment of landslide prediction models

International Nuclear Information System (INIS)

Othman, A N; Mohd, W M N W; Noraini, S

2014-01-01

The increasing population and expansion of settlements over hilly areas has greatly increased the impact of natural disasters such as landslide. Therefore, it is important to developed models which could accurately predict landslide hazard zones. Over the years, various techniques and models have been developed to predict landslide hazard zones. The aim of this paper is to access the accuracy of landslide prediction models developed by the authors. The methodology involved the selection of study area, data acquisition, data processing and model development and also data analysis. The development of these models are based on nine different landslide inducing parameters i.e. slope, land use, lithology, soil properties, geomorphology, flow accumulation, aspect, proximity to river and proximity to road. Rank sum, rating, pairwise comparison and AHP techniques are used to determine the weights for each of the parameters used. Four (4) different models which consider different parameter combinations are developed by the authors. Results obtained are compared to landslide history and accuracies for Model 1, Model 2, Model 3 and Model 4 are 66.7, 66.7%, 60% and 22.9% respectively. From the results, rank sum, rating and pairwise comparison can be useful techniques to predict landslide hazard zones

Hyperspectral-based predictive modelling of grapevine water status in the Portuguese Douro wine region

Science.gov (United States)

Pôças, Isabel; Gonçalves, João; Costa, Patrícia Malva; Gonçalves, Igor; Pereira, Luís S.; Cunha, Mario

2017-06-01

In this study, hyperspectral reflectance (HySR) data derived from a handheld spectroradiometer were used to assess the water status of three grapevine cultivars in two sub-regions of Douro wine region during two consecutive years. A large set of potential predictors derived from the HySR data were considered for modelling/predicting the predawn leaf water potential (Ψpd) through different statistical and machine learning techniques. Three HySR vegetation indices were selected as final predictors for the computation of the models and the in-season time trend was removed from data by using a time predictor. The vegetation indices selected were the Normalized Reflectance Index for the wavelengths 554 nm and 561 nm (NRI554;561), the water index (WI) for the wavelengths 900 nm and 970 nm, and the D1 index which is associated with the rate of reflectance increase in the wavelengths of 706 nm and 730 nm. These vegetation indices covered the green, red edge and the near infrared domains of the electromagnetic spectrum. A large set of state-of-the-art analysis and statistical and machine-learning modelling techniques were tested. Predictive modelling techniques based on generalized boosted model (GBM), bagged multivariate adaptive regression splines (B-MARS), generalized additive model (GAM), and Bayesian regularized neural networks (BRNN) showed the best performance for predicting Ψpd, with an average determination coefficient (R2) ranging between 0.78 and 0.80 and RMSE varying between 0.11 and 0.12 MPa. When cultivar Touriga Nacional was used for training the models and the cultivars Touriga Franca and Tinta Barroca for testing (independent validation), the models performance was good, particularly for GBM (R2 = 0.85; RMSE = 0.09 MPa). Additionally, the comparison of Ψpd observed and predicted showed an equitable dispersion of data from the various cultivars. The results achieved show a good potential of these predictive models based on vegetation indices to support

Full-Scale Approximations of Spatio-Temporal Covariance Models for Large Datasets

KAUST Repository

Zhang, Bohai

2014-01-01

Various continuously-indexed spatio-temporal process models have been constructed to characterize spatio-temporal dependence structures, but the computational complexity for model fitting and predictions grows in a cubic order with the size of dataset and application of such models is not feasible for large datasets. This article extends the full-scale approximation (FSA) approach by Sang and Huang (2012) to the spatio-temporal context to reduce computational complexity. A reversible jump Markov chain Monte Carlo (RJMCMC) algorithm is proposed to select knots automatically from a discrete set of spatio-temporal points. Our approach is applicable to nonseparable and nonstationary spatio-temporal covariance models. We illustrate the effectiveness of our method through simulation experiments and application to an ozone measurement dataset.

A Validation of Subchannel Based CHF Prediction Model for Rod Bundles

International Nuclear Information System (INIS)

Hwang, Dae-Hyun; Kim, Seong-Jin

2015-01-01

A large number of CHF data base were procured from various sources which included square and non-square lattice test bundles. CHF prediction accuracy was evaluated for various models including CHF lookup table method, empirical correlations, and phenomenological DNB models. The parametric effect of the mass velocity and unheated wall has been investigated from the experimental result, and incorporated into the development of local parameter CHF correlation applicable to APWR conditions. According to the CHF design criterion, the CHF should not occur at the hottest rod in the reactor core during normal operation and anticipated operational occurrences with at least a 95% probability at a 95% confidence level. This is accomplished by assuring that the minimum DNBR (Departure from Nucleate Boiling Ratio) in the reactor core is greater than the limit DNBR which accounts for the accuracy of CHF prediction model. The limit DNBR can be determined from the inverse of the lower tolerance limit of M/P that is evaluated from the measured-to-predicted CHF ratios for the relevant CHF data base. It is important to evaluate an adequacy of the CHF prediction model for application to the actual reactor core conditions. Validation of CHF prediction model provides the degree of accuracy inferred from the comparison of solution and data. To achieve a required accuracy for the CHF prediction model, it may be necessary to calibrate the model parameters by employing the validation results. If the accuracy of the model is acceptable, then it is applied to the real complex system with the inferred accuracy of the model. In a conventional approach, the accuracy of CHF prediction model was evaluated from the M/P statistics for relevant CHF data base, which was evaluated by comparing the nominal values of the predicted and measured CHFs. The experimental uncertainty for the CHF data was not considered in this approach to determine the limit DNBR. When a subchannel based CHF prediction model

Unified model to predict flexural shear behavior of externally bonded RC beams

International Nuclear Information System (INIS)

Colotti, V.; Spadea, G.; Swamy, R.N.

2006-01-01

Structural strengthening with externally bonded reinforcement is now recognized as a cost-effective, structurally sound and practically efficient method of rehabilitating deteriorating and damaged reinforced concrete beams. There is now an urgent need to develop a sound engineering basis which can predict the failure loads of all such strengthened beams in a reliable and consistent manner. Existing models to predict the behavior at ultimate of strengthened beams suffer from many limitations and weaknesses. This paper presents a unified global model, based on the Strut-and-Tie approach, to predict the failure loads of reinforced concrete beams strengthened for flexure and/or shear. This structural model is based on rational engineering principles, considers all the possible failure modes, and incorporates the load transfer mechanism bond to reflect the debonding phenomena which has a dominant influence on the failure process of plated beams. The model is validated against about 200 strengthened beam test reported in the literature and failing in flexure and/or shear, involving a large number of structural variables and steel, carbon and glass fiber reinforced polymer laminates as reinforcing medium. (author)

Mental models accurately predict emotion transitions.

Science.gov (United States)

Thornton, Mark A; Tamir, Diana I

2017-06-06

Successful social interactions depend on people's ability to predict others' future actions and emotions. People possess many mechanisms for perceiving others' current emotional states, but how might they use this information to predict others' future states? We hypothesized that people might capitalize on an overlooked aspect of affective experience: current emotions predict future emotions. By attending to regularities in emotion transitions, perceivers might develop accurate mental models of others' emotional dynamics. People could then use these mental models of emotion transitions to predict others' future emotions from currently observable emotions. To test this hypothesis, studies 1-3 used data from three extant experience-sampling datasets to establish the actual rates of emotional transitions. We then collected three parallel datasets in which participants rated the transition likelihoods between the same set of emotions. Participants' ratings of emotion transitions predicted others' experienced transitional likelihoods with high accuracy. Study 4 demonstrated that four conceptual dimensions of mental state representation-valence, social impact, rationality, and human mind-inform participants' mental models. Study 5 used 2 million emotion reports on the Experience Project to replicate both of these findings: again people reported accurate models of emotion transitions, and these models were informed by the same four conceptual dimensions. Importantly, neither these conceptual dimensions nor holistic similarity could fully explain participants' accuracy, suggesting that their mental models contain accurate information about emotion dynamics above and beyond what might be predicted by static emotion knowledge alone.

Mental models accurately predict emotion transitions

Science.gov (United States)

Thornton, Mark A.; Tamir, Diana I.

2017-01-01

Successful social interactions depend on people’s ability to predict others’ future actions and emotions. People possess many mechanisms for perceiving others’ current emotional states, but how might they use this information to predict others’ future states? We hypothesized that people might capitalize on an overlooked aspect of affective experience: current emotions predict future emotions. By attending to regularities in emotion transitions, perceivers might develop accurate mental models of others’ emotional dynamics. People could then use these mental models of emotion transitions to predict others’ future emotions from currently observable emotions. To test this hypothesis, studies 1–3 used data from three extant experience-sampling datasets to establish the actual rates of emotional transitions. We then collected three parallel datasets in which participants rated the transition likelihoods between the same set of emotions. Participants’ ratings of emotion transitions predicted others’ experienced transitional likelihoods with high accuracy. Study 4 demonstrated that four conceptual dimensions of mental state representation—valence, social impact, rationality, and human mind—inform participants’ mental models. Study 5 used 2 million emotion reports on the Experience Project to replicate both of these findings: again people reported accurate models of emotion transitions, and these models were informed by the same four conceptual dimensions. Importantly, neither these conceptual dimensions nor holistic similarity could fully explain participants’ accuracy, suggesting that their mental models contain accurate information about emotion dynamics above and beyond what might be predicted by static emotion knowledge alone. PMID:28533373

Poisson Mixture Regression Models for Heart Disease Prediction.

Science.gov (United States)

Mufudza, Chipo; Erol, Hamza

2016-01-01

Early heart disease control can be achieved by high disease prediction and diagnosis efficiency. This paper focuses on the use of model based clustering techniques to predict and diagnose heart disease via Poisson mixture regression models. Analysis and application of Poisson mixture regression models is here addressed under two different classes: standard and concomitant variable mixture regression models. Results show that a two-component concomitant variable Poisson mixture regression model predicts heart disease better than both the standard Poisson mixture regression model and the ordinary general linear Poisson regression model due to its low Bayesian Information Criteria value. Furthermore, a Zero Inflated Poisson Mixture Regression model turned out to be the best model for heart prediction over all models as it both clusters individuals into high or low risk category and predicts rate to heart disease componentwise given clusters available. It is deduced that heart disease prediction can be effectively done by identifying the major risks componentwise using Poisson mixture regression model.

Poisson Mixture Regression Models for Heart Disease Prediction

Science.gov (United States)

Erol, Hamza

2016-01-01

Early heart disease control can be achieved by high disease prediction and diagnosis efficiency. This paper focuses on the use of model based clustering techniques to predict and diagnose heart disease via Poisson mixture regression models. Analysis and application of Poisson mixture regression models is here addressed under two different classes: standard and concomitant variable mixture regression models. Results show that a two-component concomitant variable Poisson mixture regression model predicts heart disease better than both the standard Poisson mixture regression model and the ordinary general linear Poisson regression model due to its low Bayesian Information Criteria value. Furthermore, a Zero Inflated Poisson Mixture Regression model turned out to be the best model for heart prediction over all models as it both clusters individuals into high or low risk category and predicts rate to heart disease componentwise given clusters available. It is deduced that heart disease prediction can be effectively done by identifying the major risks componentwise using Poisson mixture regression model. PMID:27999611

Research on Fault Prediction of Distribution Network Based on Large Data

Directory of Open Access Journals (Sweden)

Jinglong Zhou

2017-01-01

Full Text Available With the continuous development of information technology and the improvement of distribution automation level. Especially, the amount of on-line monitoring and statistical data is increasing, and large data is used data distribution system, describes the technology to collect, data analysis and data processing of the data distribution system. The artificial neural network mining algorithm and the large data are researched in the fault diagnosis and prediction of the distribution network.

Predicting the evolution of large cholera outbreaks: lessons learnt from the Haiti case study

Science.gov (United States)

Bertuzzo, Enrico; Mari, Lorenzo; Righetto, Lorenzo; Knox, Allyn; Finger, Flavio; Casagrandi, Renato; Gatto, Marino; Rodriguez-Iturbe, Ignacio; Rinaldo, Andrea

2013-04-01

Mathematical models can provide key insights into the course of an ongoing epidemic, potentially aiding real-time emergency management in allocating health care resources and possibly anticipating the impact of alternative interventions. Spatially explicit models of waterborne disease are made routinely possible by widespread data mapping of hydrology, road network, population distribution, and sanitation. Here, we study the ex-post reliability of predictions of the ongoing Haiti cholera outbreak. Our model consists of a set of dynamical equations (SIR-like, i.e. subdivided into the compartments of Susceptible, Infected and Recovered individuals) describing a connected network of human communities where the infection results from the exposure to excess concentrations of pathogens in the water, which are, in turn, driven by hydrologic transport through waterways and by mobility of susceptible and infected individuals. Following the evidence of a clear correlation between rainfall events and cholera resurgence, we test a new mechanism explicitly accounting for rainfall as a driver of enhanced disease transmission by washout of open-air defecation sites or cesspool overflows. A general model for Haitian epidemic cholera and the related uncertainty is thus proposed and applied to the dataset of reported cases now available. The model allows us to draw predictions on longer-term epidemic cholera in Haiti from multi-season Monte Carlo runs, carried out up to January 2014 by using a multivariate Poisson rainfall generator, with parameters varying in space and time. Lessons learned and open issues are discussed and placed in perspective. We conclude that, despite differences in methods that can be tested through model-guided field validation, mathematical modeling of large-scale outbreaks emerges as an essential component of future cholera epidemic control.

Comparisons of Faulting-Based Pavement Performance Prediction Models

Directory of Open Access Journals (Sweden)

Weina Wang

2017-01-01

Full Text Available Faulting prediction is the core of concrete pavement maintenance and design. Highway agencies are always faced with the problem of lower accuracy for the prediction which causes costly maintenance. Although many researchers have developed some performance prediction models, the accuracy of prediction has remained a challenge. This paper reviews performance prediction models and JPCP faulting models that have been used in past research. Then three models including multivariate nonlinear regression (MNLR model, artificial neural network (ANN model, and Markov Chain (MC model are tested and compared using a set of actual pavement survey data taken on interstate highway with varying design features, traffic, and climate data. It is found that MNLR model needs further recalibration, while the ANN model needs more data for training the network. MC model seems a good tool for pavement performance prediction when the data is limited, but it is based on visual inspections and not explicitly related to quantitative physical parameters. This paper then suggests that the further direction for developing the performance prediction model is incorporating the advantages and disadvantages of different models to obtain better accuracy.

Unreachable Setpoints in Model Predictive Control

DEFF Research Database (Denmark)

Rawlings, James B.; Bonné, Dennis; Jørgensen, John Bagterp

2008-01-01

In this work, a new model predictive controller is developed that handles unreachable setpoints better than traditional model predictive control methods. The new controller induces an interesting fast/slow asymmetry in the tracking response of the system. Nominal asymptotic stability of the optimal...... steady state is established for terminal constraint model predictive control (MPC). The region of attraction is the steerable set. Existing analysis methods for closed-loop properties of MPC are not applicable to this new formulation, and a new analysis method is developed. It is shown how to extend...

On the Bayesian calibration of computer model mixtures through experimental data, and the design of predictive models

Science.gov (United States)

Karagiannis, Georgios; Lin, Guang

2017-08-01

For many real systems, several computer models may exist with different physics and predictive abilities. To achieve more accurate simulations/predictions, it is desirable for these models to be properly combined and calibrated. We propose the Bayesian calibration of computer model mixture method which relies on the idea of representing the real system output as a mixture of the available computer model outputs with unknown input dependent weight functions. The method builds a fully Bayesian predictive model as an emulator for the real system output by combining, weighting, and calibrating the available models in the Bayesian framework. Moreover, it fits a mixture of calibrated computer models that can be used by the domain scientist as a mean to combine the available computer models, in a flexible and principled manner, and perform reliable simulations. It can address realistic cases where one model may be more accurate than the others at different input values because the mixture weights, indicating the contribution of each model, are functions of the input. Inference on the calibration parameters can consider multiple computer models associated with different physics. The method does not require knowledge of the fidelity order of the models. We provide a technique able to mitigate the computational overhead due to the consideration of multiple computer models that is suitable to the mixture model framework. We implement the proposed method in a real-world application involving the Weather Research and Forecasting large-scale climate model.

Learning Traffic as Images: A Deep Convolutional Neural Network for Large-Scale Transportation Network Speed Prediction.

Science.gov (United States)

Ma, Xiaolei; Dai, Zhuang; He, Zhengbing; Ma, Jihui; Wang, Yong; Wang, Yunpeng

2017-04-10

This paper proposes a convolutional neural network (CNN)-based method that learns traffic as images and predicts large-scale, network-wide traffic speed with a high accuracy. Spatiotemporal traffic dynamics are converted to images describing the time and space relations of traffic flow via a two-dimensional time-space matrix. A CNN is applied to the image following two consecutive steps: abstract traffic feature extraction and network-wide traffic speed prediction. The effectiveness of the proposed method is evaluated by taking two real-world transportation networks, the second ring road and north-east transportation network in Beijing, as examples, and comparing the method with four prevailing algorithms, namely, ordinary least squares, k-nearest neighbors, artificial neural network, and random forest, and three deep learning architectures, namely, stacked autoencoder, recurrent neural network, and long-short-term memory network. The results show that the proposed method outperforms other algorithms by an average accuracy improvement of 42.91% within an acceptable execution time. The CNN can train the model in a reasonable time and, thus, is suitable for large-scale transportation networks.

A regional neural network model for predicting mean daily river water temperature

Science.gov (United States)

Wagner, Tyler; DeWeber, Jefferson Tyrell

2014-01-01

Water temperature is a fundamental property of river habitat and often a key aspect of river resource management, but measurements to characterize thermal regimes are not available for most streams and rivers. As such, we developed an artificial neural network (ANN) ensemble model to predict mean daily water temperature in 197,402 individual stream reaches during the warm season (May–October) throughout the native range of brook trout Salvelinus fontinalis in the eastern U.S. We compared four models with different groups of predictors to determine how well water temperature could be predicted by climatic, landform, and land cover attributes, and used the median prediction from an ensemble of 100 ANNs as our final prediction for each model. The final model included air temperature, landform attributes and forested land cover and predicted mean daily water temperatures with moderate accuracy as determined by root mean squared error (RMSE) at 886 training sites with data from 1980 to 2009 (RMSE = 1.91 °C). Based on validation at 96 sites (RMSE = 1.82) and separately for data from 2010 (RMSE = 1.93), a year with relatively warmer conditions, the model was able to generalize to new stream reaches and years. The most important predictors were mean daily air temperature, prior 7 day mean air temperature, and network catchment area according to sensitivity analyses. Forest land cover at both riparian and catchment extents had relatively weak but clear negative effects. Predicted daily water temperature averaged for the month of July matched expected spatial trends with cooler temperatures in headwaters and at higher elevations and latitudes. Our ANN ensemble is unique in predicting daily temperatures throughout a large region, while other regional efforts have predicted at relatively coarse time steps. The model may prove a useful tool for predicting water temperatures in sampled and unsampled rivers under current conditions and future projections of climate

Analytical model for local scour prediction around hydrokinetic turbine foundations

Science.gov (United States)

Musa, M.; Heisel, M.; Hill, C.; Guala, M.

2017-12-01

Marine and Hydrokinetic renewable energy is an emerging sustainable and secure technology which produces clean energy harnessing water currents from mostly tidal and fluvial waterways. Hydrokinetic turbines are typically anchored at the bottom of the channel, which can be erodible or non-erodible. Recent experiments demonstrated the interactions between operating turbines and an erodible surface with sediment transport, resulting in a remarkable localized erosion-deposition pattern significantly larger than those observed by static in-river construction such as bridge piers, etc. Predicting local scour geometry at the base of hydrokinetic devices is extremely important during foundation design, installation, operation, and maintenance (IO&M), and long-term structural integrity. An analytical modeling framework is proposed applying the phenomenological theory of turbulence to the flow structures that promote the scouring process at the base of a turbine. The evolution of scour is directly linked to device operating conditions through the turbine drag force, which is inferred to locally dictate the energy dissipation rate in the scour region. The predictive model is validated using experimental data obtained at the University of Minnesota's St. Anthony Falls Laboratory (SAFL), covering two sediment mobility regimes (clear water and live bed), different turbine designs, hydraulic parameters, grain size distribution and bedform types. The model is applied to a potential prototype scale deployment in the lower Mississippi River, demonstrating its practical relevance and endorsing the feasibility of hydrokinetic energy power plants in large sandy rivers. Multi-turbine deployments are further studied experimentally by monitoring both local and non-local geomorphic effects introduced by a twelve turbine staggered array model installed in a wide channel at SAFL. Local scour behind each turbine is well captured by the theoretical predictive model. However, multi

Predictive modeling of reactive wetting and metal joining.

Energy Technology Data Exchange (ETDEWEB)

van Swol, Frank B.

2013-09-01

The performance, reproducibility and reliability of metal joints are complex functions of the detailed history of physical processes involved in their creation. Prediction and control of these processes constitutes an intrinsically challenging multi-physics problem involving heating and melting a metal alloy and reactive wetting. Understanding this process requires coupling strong molecularscale chemistry at the interface with microscopic (diffusion) and macroscopic mass transport (flow) inside the liquid followed by subsequent cooling and solidification of the new metal mixture. The final joint displays compositional heterogeneity and its resulting microstructure largely determines the success or failure of the entire component. At present there exists no computational tool at Sandia that can predict the formation and success of a braze joint, as current capabilities lack the ability to capture surface/interface reactions and their effect on interface properties. This situation precludes us from implementing a proactive strategy to deal with joining problems. Here, we describe what is needed to arrive at a predictive modeling and simulation capability for multicomponent metals with complicated phase diagrams for melting and solidification, incorporating dissolutive and composition-dependent wetting.

A systematic review of breast cancer incidence risk prediction models with meta-analysis of their performance.

Science.gov (United States)

Meads, Catherine; Ahmed, Ikhlaaq; Riley, Richard D

2012-04-01

'Gail 2' model showed the average C statistic was 0.63 (95% CI 0.59-0.67), and the expected/observed ratio of events varied considerably across studies (95% prediction interval for E/O ratio when the model was applied in practice was 0.75-1.19). There is a need for models with better predictive performance but, given the large amount of work already conducted, further improvement of existing models based on conventional risk factors is perhaps unlikely. Research to identify new risk factors with large additionally predictive ability is therefore needed, alongside clearer reporting and continual validation of new models as they develop.

An adaptive two-stage analog/regression model for probabilistic prediction of small-scale precipitation in France

Directory of Open Access Journals (Sweden)

J. Chardon

2018-01-01

Full Text Available Statistical downscaling models (SDMs are often used to produce local weather scenarios from large-scale atmospheric information. SDMs include transfer functions which are based on a statistical link identified from observations between local weather and a set of large-scale predictors. As physical processes driving surface weather vary in time, the most relevant predictors and the regression link are likely to vary in time too. This is well known for precipitation for instance and the link is thus often estimated after some seasonal stratification of the data. In this study, we present a two-stage analog/regression model where the regression link is estimated from atmospheric analogs of the current prediction day. Atmospheric analogs are identified from fields of geopotential heights at 1000 and 500 hPa. For the regression stage, two generalized linear models are further used to model the probability of precipitation occurrence and the distribution of non-zero precipitation amounts, respectively. The two-stage model is evaluated for the probabilistic prediction of small-scale precipitation over France. It noticeably improves the skill of the prediction for both precipitation occurrence and amount. As the analog days vary from one prediction day to another, the atmospheric predictors selected in the regression stage and the value of the corresponding regression coefficients can vary from one prediction day to another. The model allows thus for a day-to-day adaptive and tailored downscaling. It can also reveal specific predictors for peculiar and non-frequent weather configurations.

Meta-analysis of choice set generation effects on route choice model estimates and predictions

DEFF Research Database (Denmark)

Prato, Carlo Giacomo

2012-01-01

are applied for model estimation and results are compared to the ‘true model estimates’. Last, predictions from the simulation of models estimated with objective choice sets are compared to the ‘postulated predicted routes’. A meta-analytical approach allows synthesizing the effect of judgments......Large scale applications of behaviorally realistic transport models pose several challenges to transport modelers on both the demand and the supply sides. On the supply side, path-based solutions to the user assignment equilibrium problem help modelers in enhancing the route choice behavior...... modeling, but require them to generate choice sets by selecting a path generation technique and its parameters according to personal judgments. This paper proposes a methodology and an experimental setting to provide general indications about objective judgments for an effective route choice set generation...

Estimating Model Prediction Error: Should You Treat Predictions as Fixed or Random?

Science.gov (United States)

Wallach, Daniel; Thorburn, Peter; Asseng, Senthold; Challinor, Andrew J.; Ewert, Frank; Jones, James W.; Rotter, Reimund; Ruane, Alexander

2016-01-01

Crop models are important tools for impact assessment of climate change, as well as for exploring management options under current climate. It is essential to evaluate the uncertainty associated with predictions of these models. We compare two criteria of prediction error; MSEP fixed, which evaluates mean squared error of prediction for a model with fixed structure, parameters and inputs, and MSEP uncertain( X), which evaluates mean squared error averaged over the distributions of model structure, inputs and parameters. Comparison of model outputs with data can be used to estimate the former. The latter has a squared bias term, which can be estimated using hindcasts, and a model variance term, which can be estimated from a simulation experiment. The separate contributions to MSEP uncertain (X) can be estimated using a random effects ANOVA. It is argued that MSEP uncertain (X) is the more informative uncertainty criterion, because it is specific to each prediction situation.

Risk terrain modeling predicts child maltreatment.

Science.gov (United States)

Daley, Dyann; Bachmann, Michael; Bachmann, Brittany A; Pedigo, Christian; Bui, Minh-Thuy; Coffman, Jamye

2016-12-01

As indicated by research on the long-term effects of adverse childhood experiences (ACEs), maltreatment has far-reaching consequences for affected children. Effective prevention measures have been elusive, partly due to difficulty in identifying vulnerable children before they are harmed. This study employs Risk Terrain Modeling (RTM), an analysis of the cumulative effect of environmental factors thought to be conducive for child maltreatment, to create a highly accurate prediction model for future substantiated child maltreatment cases in the City of Fort Worth, Texas. The model is superior to commonly used hotspot predictions and more beneficial in aiding prevention efforts in a number of ways: 1) it identifies the highest risk areas for future instances of child maltreatment with improved precision and accuracy; 2) it aids the prioritization of risk-mitigating efforts by informing about the relative importance of the most significant contributing risk factors; 3) since predictions are modeled as a function of easily obtainable data, practitioners do not have to undergo the difficult process of obtaining official child maltreatment data to apply it; 4) the inclusion of a multitude of environmental risk factors creates a more robust model with higher predictive validity; and, 5) the model does not rely on a retrospective examination of past instances of child maltreatment, but adapts predictions to changing environmental conditions. The present study introduces and examines the predictive power of this new tool to aid prevention efforts seeking to improve the safety, health, and wellbeing of vulnerable children. Copyright © 2016 The Authors. Published by Elsevier Ltd.. All rights reserved.

Prediction of crack propagation and arrest in X100 natural gas transmission pipelines with a strain rate dependent damage model (SRDD). Part 2: Large scale pipe models with gas depressurisation

International Nuclear Information System (INIS)

Oikonomidis, F.; Shterenlikht, A.; Truman, C.E.

2014-01-01

Part 1 of this paper described a specimen for the measurement of high strain rate flow and fracture properties of pipe material and for tuning a strain rate dependent damage model (SRDD). In part 2 the tuned SRDD model is used for the simulation of axial crack propagation and arrest in X100 natural gas pipelines. Linear pressure drop model was adopted behind the crack tip, and an exponential gas depressurisation model was used ahead of the crack tip. The model correctly predicted the crack initiation (burst) pressure, the crack speed and the crack arrest length. Strain rates between 1000 s −1 and 3000 s −1 immediately ahead of the crack tip are predicted, giving a strong indication that a strain rate material model is required for the structural integrity assessment of the natural gas pipelines. The models predict the stress triaxiality of about 0.65 for at least 1 m ahead of the crack tip, gradually dropping to 0.5 at distances of about 5–7 m ahead of the crack tip. Finally, the models predicted a linear drop in crack tip opening angle (CTOA) from about 11−12° at the onset of crack propagation down to 7−8° at crack arrest. Only the lower of these values agree with those reported in the literature for quasi-static measurements. This discrepancy might indicate substantial strain rate dependence in CTOA. - Highlights: • Finite element simulations of 3 burst tests of X100 pipes are detailed. • Strain rate dependent damage model, tuned on small scale X100 samples, was used. • The models correctly predict burst pressure, crack speed and crack arrest length. • The model predicts a crack length dependent critical CTOA. • The strain rate dependent damage model is verified as mesh independent

Case studies in archaeological predictive modelling

NARCIS (Netherlands)

Verhagen, Jacobus Wilhelmus Hermanus Philippus

2007-01-01

In this thesis, a collection of papers is put together dealing with various quantitative aspects of predictive modelling and archaeological prospection. Among the issues covered are the effects of survey bias on the archaeological data used for predictive modelling, and the complexities of testing

Receiver Operating Characteristic Curve-Based Prediction Model for Periodontal Disease Updated With the Calibrated Community Periodontal Index.

Science.gov (United States)

Su, Chiu-Wen; Yen, Amy Ming-Fang; Lai, Hongmin; Chen, Hsiu-Hsi; Chen, Sam Li-Sheng

2017-12-01

The accuracy of a prediction model for periodontal disease using the community periodontal index (CPI) has been undertaken by using an area under a receiver operating characteristics (AUROC) curve. How the uncalibrated CPI, as measured by general dentists trained by periodontists in a large epidemiologic study, and affects the performance in a prediction model, has not been researched yet. A two-stage design was conducted by first proposing a validation study to calibrate CPI between a senior periodontal specialist and trained general dentists who measured CPIs in the main study of a nationwide survey. A Bayesian hierarchical logistic regression model was applied to estimate the non-updated and updated clinical weights used for building up risk scores. How the calibrated CPI affected performance of the updated prediction model was quantified by comparing AUROC curves between the original and updated models. Estimates regarding calibration of CPI obtained from the validation study were 66% and 85% for sensitivity and specificity, respectively. After updating, clinical weights of each predictor were inflated, and the risk score for the highest risk category was elevated from 434 to 630. Such an update improved the AUROC performance of the two corresponding prediction models from 62.6% (95% confidence interval [CI]: 61.7% to 63.6%) for the non-updated model to 68.9% (95% CI: 68.0% to 69.6%) for the updated one, reaching a statistically significant difference (P prediction model was demonstrated for periodontal disease as measured by the calibrated CPI derived from a large epidemiologic survey.

Development of Shear Capacity Prediction Model for FRP-RC Beam without Web Reinforcement

Directory of Open Access Journals (Sweden)

Md. Arman Chowdhury

2016-01-01

Full Text Available Available codes and models generally use partially modified shear design equation, developed earlier for steel reinforced concrete, for predicting the shear capacity of FRP-RC members. Consequently, calculated shear capacity shows under- or overestimation. Furthermore, in most models some affecting parameters of shear strength are overlooked. In this study, a new and simplified shear capacity prediction model is proposed considering all the parameters. A large database containing 157 experimental results of FRP-RC beams without shear reinforcement is assembled from the published literature. A parametric study is then performed to verify the accuracy of the proposed model. Again, a comprehensive review of 9 codes and 12 available models is done, published back from 1997 to date for comparison with the proposed model. Hence, it is observed that the proposed equation shows overall optimized performance compared to all the codes and models within the range of used experimental dataset.

Wall modeled large eddy simulations of complex high Reynolds number flows with synthetic inlet turbulence

International Nuclear Information System (INIS)

Patil, Sunil; Tafti, Danesh

2012-01-01

Highlights: ► Large eddy simulation. ► Wall layer modeling. ► Synthetic inlet turbulence. ► Swirl flows. - Abstract: Large eddy simulations of complex high Reynolds number flows are carried out with the near wall region being modeled with a zonal two layer model. A novel formulation for solving the turbulent boundary layer equation for the effective tangential velocity in a generalized co-ordinate system is presented and applied in the near wall zonal treatment. This formulation reduces the computational time in the inner layer significantly compared to the conventional two layer formulations present in the literature and is most suitable for complex geometries involving body fitted structured and unstructured meshes. The cost effectiveness and accuracy of the proposed wall model, used with the synthetic eddy method (SEM) to generate inlet turbulence, is investigated in turbulent channel flow, flow over a backward facing step, and confined swirling flows at moderately high Reynolds numbers. Predictions are compared with available DNS, experimental LDV data, as well as wall resolved LES. In all cases, there is at least an order of magnitude reduction in computational cost with no significant loss in prediction accuracy.

Structure of exotic nuclei by large-scale shell model calculations

International Nuclear Information System (INIS)

Utsuno, Yutaka; Otsuka, Takaharu; Mizusaki, Takahiro; Honma, Michio

2006-01-01

An extensive large-scale shell-model study is conducted for unstable nuclei around N = 20 and N = 28, aiming to investigate how the shell structure evolves from stable to unstable nuclei and affects the nuclear structure. The structure around N = 20 including the disappearance of the magic number is reproduced systematically, exemplified in the systematics of the electromagnetic moments in the Na isotope chain. As a key ingredient dominating the structure/shell evolution in the exotic nuclei from a general viewpoint, we pay attention to the tensor force. Including a proper strength of the tensor force in the effective interaction, we successfully reproduce the proton shell evolution ranging from N = 20 to 28 without any arbitrary modifications in the interaction and predict the ground state of 42Si to contain a large deformed component

Application of model predictive control for optimal operation of wind turbines

Science.gov (United States)

Yuan, Yuan; Cao, Pei; Tang, J.

2017-04-01

For large-scale wind turbines, reducing maintenance cost is a major challenge. Model predictive control (MPC) is a promising approach to deal with multiple conflicting objectives using the weighed sum approach. In this research, model predictive control method is applied to wind turbine to find an optimal balance between multiple objectives, such as the energy capture, loads on turbine components, and the pitch actuator usage. The actuator constraints are integrated into the objective function at the control design stage. The analysis is carried out in both the partial load region and full load region, and the performances are compared with those of a baseline gain scheduling PID controller. The application of this strategy achieves enhanced balance of component loads, the average power and actuator usages in partial load region.

Fingerprint verification prediction model in hand dermatitis.

Science.gov (United States)

Lee, Chew K; Chang, Choong C; Johor, Asmah; Othman, Puwira; Baba, Roshidah

2015-07-01

Hand dermatitis associated fingerprint changes is a significant problem and affects fingerprint verification processes. This study was done to develop a clinically useful prediction model for fingerprint verification in patients with hand dermatitis. A case-control study involving 100 patients with hand dermatitis. All patients verified their thumbprints against their identity card. Registered fingerprints were randomized into a model derivation and model validation group. Predictive model was derived using multiple logistic regression. Validation was done using the goodness-of-fit test. The fingerprint verification prediction model consists of a major criterion (fingerprint dystrophy area of ≥ 25%) and two minor criteria (long horizontal lines and long vertical lines). The presence of the major criterion predicts it will almost always fail verification, while presence of both minor criteria and presence of one minor criterion predict high and low risk of fingerprint verification failure, respectively. When none of the criteria are met, the fingerprint almost always passes the verification. The area under the receiver operating characteristic curve was 0.937, and the goodness-of-fit test showed agreement between the observed and expected number (P = 0.26). The derived fingerprint verification failure prediction model is validated and highly discriminatory in predicting risk of fingerprint verification in patients with hand dermatitis. © 2014 The International Society of Dermatology.

Evaluation of sub grid scale and local wall models in Large-eddy simulations of separated flow

Directory of Open Access Journals (Sweden)

Sam Ali Al

2015-01-01

Full Text Available The performance of the Sub Grid Scale models is studied by simulating a separated flow over a wavy channel. The first and second order statistical moments of the resolved velocities obtained by using Large-Eddy simulations at different mesh resolutions are compared with Direct Numerical Simulations data. The effectiveness of modeling the wall stresses by using local log-law is then tested on a relatively coarse grid. The results exhibit a good agreement between highly-resolved Large Eddy Simulations and Direct Numerical Simulations data regardless the Sub Grid Scale models. However, the agreement is less satisfactory with relatively coarse grid without using any wall models and the differences between Sub Grid Scale models are distinguishable. Using local wall model retuned the basic flow topology and reduced significantly the differences between the coarse meshed Large-Eddy Simulations and Direct Numerical Simulations data. The results show that the ability of local wall model to predict the separation zone depends strongly on its implementation way.

Risk Prediction Models in Psychiatry: Toward a New Frontier for the Prevention of Mental Illnesses.

Science.gov (United States)

Bernardini, Francesco; Attademo, Luigi; Cleary, Sean D; Luther, Charles; Shim, Ruth S; Quartesan, Roberto; Compton, Michael T

2017-05-01

We conducted a systematic, qualitative review of risk prediction models designed and tested for depression, bipolar disorder, generalized anxiety disorder, posttraumatic stress disorder, and psychotic disorders. Our aim was to understand the current state of research on risk prediction models for these 5 disorders and thus future directions as our field moves toward embracing prediction and prevention. Systematic searches of the entire MEDLINE electronic database were conducted independently by 2 of the authors (from 1960 through 2013) in July 2014 using defined search criteria. Search terms included risk prediction, predictive model, or prediction model combined with depression, bipolar, manic depressive, generalized anxiety, posttraumatic, PTSD, schizophrenia, or psychosis. We identified 268 articles based on the search terms and 3 criteria: published in English, provided empirical data (as opposed to review articles), and presented results pertaining to developing or validating a risk prediction model in which the outcome was the diagnosis of 1 of the 5 aforementioned mental illnesses. We selected 43 original research reports as a final set of articles to be qualitatively reviewed. The 2 independent reviewers abstracted 3 types of data (sample characteristics, variables included in the model, and reported model statistics) and reached consensus regarding any discrepant abstracted information. Twelve reports described models developed for prediction of major depressive disorder, 1 for bipolar disorder, 2 for generalized anxiety disorder, 4 for posttraumatic stress disorder, and 24 for psychotic disorders. Most studies reported on sensitivity, specificity, positive predictive value, negative predictive value, and area under the (receiver operating characteristic) curve. Recent studies demonstrate the feasibility of developing risk prediction models for psychiatric disorders (especially psychotic disorders). The field must now advance by (1) conducting more large

Improved Predictions of the Geographic Distribution of Invasive Plants Using Climatic Niche Models

Science.gov (United States)

Ramírez-Albores, Jorge E.; Bustamante, Ramiro O.

2016-01-01

Climatic niche models for invasive plants are usually constructed with occurrence records taken from literature and collections. Because these data neither discriminate among life-cycle stages of plants (adult or juvenile) nor the origin of individuals (naturally established or man-planted), the resulting models may mispredict the distribution ranges of these species. We propose that more accurate predictions could be obtained by modelling climatic niches with data of naturally established individuals, particularly with occurrence records of juvenile plants because this would restrict the predictions of models to those sites where climatic conditions allow the recruitment of the species. To test this proposal, we focused on the Peruvian peppertree (Schinus molle), a South American species that has largely invaded Mexico. Three climatic niche models were constructed for this species using high-resolution dataset gathered in the field. The first model included all occurrence records, irrespective of the life-cycle stage or origin of peppertrees (generalized niche model). The second model only included occurrence records of naturally established mature individuals (adult niche model), while the third model was constructed with occurrence records of naturally established juvenile plants (regeneration niche model). When models were compared, the generalized climatic niche model predicted the presence of peppertrees in sites located farther beyond the climatic thresholds that naturally established individuals can tolerate, suggesting that human activities influence the distribution of this invasive species. The adult and regeneration climatic niche models concurred in their predictions about the distribution of peppertrees, suggesting that naturally established adult trees only occur in sites where climatic conditions allow the recruitment of juvenile stages. These results support the proposal that climatic niches of invasive plants should be modelled with data of

Finding Furfural Hydrogenation Catalysts via Predictive Modelling.

Science.gov (United States)

Strassberger, Zea; Mooijman, Maurice; Ruijter, Eelco; Alberts, Albert H; Maldonado, Ana G; Orru, Romano V A; Rothenberg, Gadi

2010-09-10

We combine multicomponent reactions, catalytic performance studies and predictive modelling to find transfer hydrogenation catalysts. An initial set of 18 ruthenium-carbene complexes were synthesized and screened in the transfer hydrogenation of furfural to furfurol with isopropyl alcohol complexes gave varied yields, from 62% up to >99.9%, with no obvious structure/activity correlations. Control experiments proved that the carbene ligand remains coordinated to the ruthenium centre throughout the reaction. Deuterium-labelling studies showed a secondary isotope effect (k(H):k(D)=1.5). Further mechanistic studies showed that this transfer hydrogenation follows the so-called monohydride pathway. Using these data, we built a predictive model for 13 of the catalysts, based on 2D and 3D molecular descriptors. We tested and validated the model using the remaining five catalysts (cross-validation, R(2)=0.913). Then, with this model, the conversion and selectivity were predicted for four completely new ruthenium-carbene complexes. These four catalysts were then synthesized and tested. The results were within 3% of the model's predictions, demonstrating the validity and value of predictive modelling in catalyst optimization.

Model Predictive Control for Smart Energy Systems

DEFF Research Database (Denmark)

Halvgaard, Rasmus

pumps, heat tanks, electrical vehicle battery charging/discharging, wind farms, power plants). 2.Embed forecasting methodologies for the weather (e.g. temperature, solar radiation), the electricity consumption, and the electricity price in a predictive control system. 3.Develop optimization algorithms....... Chapter 3 introduces Model Predictive Control (MPC) including state estimation, filtering and prediction for linear models. Chapter 4 simulates the models from Chapter 2 with the certainty equivalent MPC from Chapter 3. An economic MPC minimizes the costs of consumption based on real electricity prices...... that determined the flexibility of the units. A predictive control system easily handles constraints, e.g. limitations in power consumption, and predicts the future behavior of a unit by integrating predictions of electricity prices, consumption, and weather variables. The simulations demonstrate the expected...

The Phyre2 web portal for protein modeling, prediction and analysis.

Science.gov (United States)

Kelley, Lawrence A; Mezulis, Stefans; Yates, Christopher M; Wass, Mark N; Sternberg, Michael J E

2015-06-01

Phyre2 is a suite of tools available on the web to predict and analyze protein structure, function and mutations. The focus of Phyre2 is to provide biologists with a simple and intuitive interface to state-of-the-art protein bioinformatics tools. Phyre2 replaces Phyre, the original version of the server for which we previously published a paper in Nature Protocols. In this updated protocol, we describe Phyre2, which uses advanced remote homology detection methods to build 3D models, predict ligand binding sites and analyze the effect of amino acid variants (e.g., nonsynonymous SNPs (nsSNPs)) for a user's protein sequence. Users are guided through results by a simple interface at a level of detail they determine. This protocol will guide users from submitting a protein sequence to interpreting the secondary and tertiary structure of their models, their domain composition and model quality. A range of additional available tools is described to find a protein structure in a genome, to submit large number of sequences at once and to automatically run weekly searches for proteins that are difficult to model. The server is available at http://www.sbg.bio.ic.ac.uk/phyre2. A typical structure prediction will be returned between 30 min and 2 h after submission.

Large Civil Tiltrotor (LCTR2) Interior Noise Predictions due to Turbulent Boundary Layer Excitation

Science.gov (United States)

Grosveld, Ferdinand W.

2013-01-01

The Large Civil Tiltrotor (LCTR2) is a conceptual vehicle that has a design goal to transport 90 passengers over a distance of 1800 km at a speed of 556 km/hr. In this study noise predictions were made in the notional LCTR2 cabin due to Cockburn/Robertson and Efimtsov turbulent boundary layer (TBL) excitation models. A narrowband hybrid Finite Element (FE) analysis was performed for the low frequencies (6-141 Hz) and a Statistical Energy Analysis (SEA) was conducted for the high frequency one-third octave bands (125- 8000 Hz). It is shown that the interior sound pressure level distribution in the low frequencies is governed by interactions between individual structural and acoustic modes. The spatially averaged predicted interior sound pressure levels for the low frequency hybrid FE and the high frequency SEA analyses, due to the Efimtsov turbulent boundary layer excitation, were within 1 dB in the common 125 Hz one-third octave band. The averaged interior noise levels for the LCTR2 cabin were predicted lower than the levels in a comparable Bombardier Q400 aircraft cabin during cruise flight due to the higher cruise altitude and lower Mach number of the LCTR2. LCTR2 cabin noise due to TBL excitation during cruise flight was found not unacceptable for crew or passengers when predictions were compared to an acoustic survey on a Q400 aircraft.

Resin infusion of large composite structures modeling and manufacturing process

Energy Technology Data Exchange (ETDEWEB)

Loos, A.C. [Michigan State Univ., Dept. of Mechanical Engineering, East Lansing, MI (United States)

2006-07-01

The resin infusion processes resin transfer molding (RTM), resin film infusion (RFI) and vacuum assisted resin transfer molding (VARTM) are cost effective techniques for the fabrication of complex shaped composite structures. The dry fibrous preform is placed in the mold, consolidated, resin impregnated and cured in a single step process. The fibrous performs are often constructed near net shape using highly automated textile processes such as knitting, weaving and braiding. In this paper, the infusion processes RTM, RFI and VARTM are discussed along with the advantages of each technique compared with traditional composite fabrication methods such as prepreg tape lay up and autoclave cure. The large number of processing variables and the complex material behavior during infiltration and cure make experimental optimization of the infusion processes costly and inefficient. Numerical models have been developed which can be used to simulate the resin infusion processes. The model formulation and solution procedures for the VARTM process are presented. A VARTM process simulation of a carbon fiber preform was presented to demonstrate the type of information that can be generated by the model and to compare the model predictions with experimental measurements. Overall, the predicted flow front positions, resin pressures and preform thicknesses agree well with the measured values. The results of the simulation show the potential cost and performance benefits that can be realized by using a simulation model as part of the development process. (au)

Design and modelling of innovative machinery systems for large ships

DEFF Research Database (Denmark)

Larsen, Ulrik

Eighty percent of the growing global merchandise trade is transported by sea. The shipping industry is required to reduce the pollution and increase the energy efficiency of ships in the near future. There is a relatively large potential for approaching these requirements by implementing waste heat...... consisting of a two-zone combustion and NOx emission model, a double Wiebe heat release model, the Redlich-Kwong equation of state and the Woschni heat loss correlation. A novel methodology is presented and used to determine the optimum organic Rankine cycle process layout, working fluid and process......, are evaluated with regards to the fuel consumption and NOx emissions trade-off. The results of the calibration and validation of the engine model suggest that the main performance parameters can be predicted with adequate accuracies for the overall purpose. The results of the ORC and the Kalina cycle...

Predicting capacities of runways serving new large aircraft

Directory of Open Access Journals (Sweden)

K. Gopalakrishnan

2008-03-01

Full Text Available This paper presents a simplified approach for predicting the allowable load repetitions of New Large Aircraft (NLA loading for airfield runways based on Non-Destructive Test (NDT data. Full-scale traffic test results from the Federal Aviation Administration’s National Airport Pavement Test Facility (NAPTF were used to develop the NDT-based evaluation methodology. Four flexible test pavement sections with variable (unbound layer thicknesses were trafficked using six-wheel and four-wheel NLA test gears until the test pavements were deemed failed. Non-destructive tests using a Heavy Weight Deflectometer (HWD were conducted prior to the initiation of traffic testing to measure the pavement surface deflections. In the past, pavement surface deflections have been successfully used as an indicator of airport pavement life. In this study, the HWD surface deflections and the derived Deflection Basin Parameters (DBPs were related to functional performance of NAPTF flexible pavements through simple regression analysis. The results demonstrated the usefulness of NDT data for predicting the performance of airport flexible pavements serving the next generation of aircrafts.

High-fidelity large eddy simulation for supersonic jet noise prediction

Science.gov (United States)

Aikens, Kurt M.

The problem of intense sound radiation from supersonic jets is a concern for both civil and military applications. As a result, many experimental and computational efforts are focused at evaluating possible noise suppression techniques. Large-eddy simulation (LES) is utilized in many computational studies to simulate the turbulent jet flowfield. Integral methods such as the Ffowcs Williams-Hawkings (FWH) method are then used for propagation of the sound waves to the farfield. Improving the accuracy of this two-step methodology and evaluating beveled converging-diverging nozzles for noise suppression are the main tasks of this work. First, a series of numerical experiments are undertaken to ensure adequate numerical accuracy of the FWH methodology. This includes an analysis of different treatments for the downstream integration surface: with or without including an end-cap, averaging over multiple end-caps, and including an approximate surface integral correction term. Secondly, shock-capturing methods based on characteristic filtering and adaptive spatial filtering are used to extend a highly-parallelizable multiblock subsonic LES code to enable simulations of supersonic jets. The code is based on high-order numerical methods for accurate prediction of the acoustic sources and propagation of the sound waves. Furthermore, this new code is more efficient than the legacy version, allows cylindrical multiblock topologies, and is capable of simulating nozzles with resolved turbulent boundary layers when coupled with an approximate turbulent inflow boundary condition. Even though such wall-resolved simulations are more physically accurate, their expense is often prohibitive. To make simulations more economical, a wall model is developed and implemented. The wall modeling methodology is validated for turbulent quasi-incompressible and compressible zero pressure gradient flat plate boundary layers, and for subsonic and supersonic jets. The supersonic code additions and the

Recent Successes and Remaining Challenges in Predicting Phosphorus Loading to Surface Waters at Large Scales

Science.gov (United States)

Harrison, J.; Metson, G.; Beusen, A.

2017-12-01

Over the past century humans have greatly accelerated phosphorus (P) flows from land to aquatic ecosystems, causing eutrophication and associated effects such as harmful algal blooms and hypoxia. Effectively addressing this challenge requires understanding geographic and temporal distribution of aquatic P loading, knowledge of major controls on P loading, and the relative importance of various potential P sources. The Global (N)utrient (E)xport from (W)ater(S)heds) NEWS model and recent improvements and extensions of this modeling system can be used to generate this understanding. This presentation will focus on insights global NEWS models grant into past, present, and potential future P sources and sinks, with a focus on the world's large rivers. Early results suggest: 1) that while aquatic P loading is globally dominated by particulate forms, dissolved P can be locally dominant; 2) that P loading has increased substantially at the global scale, but unevenly between world regions, with hotspots in South and East Asia; 3) that P loading is likely to continue to increase globally, but decrease in certain regions that are actively pursuing proactive P management; and 4) that point sources, especially in urban centers, play an important (even dominant) role in determining loads of dissolved inorganic P. Despite these insights, substantial unexplained variance remains when model predictions and measurements are compared at global and regional scales, for example within the U.S. Disagreements between model predictions and measurements suggest opportunities for model improvement. In particular, explicit inclusion of soil characteristics and the concept of temporal P legacies in future iterations of NEWS (and other) models may help improve correspondence between models and measurements.

Benchmarking Deep Learning Models on Large Healthcare Datasets.

Science.gov (United States)

Purushotham, Sanjay; Meng, Chuizheng; Che, Zhengping; Liu, Yan

2018-06-04

Deep learning models (aka Deep Neural Networks) have revolutionized many fields including computer vision, natural language processing, speech recognition, and is being increasingly used in clinical healthcare applications. However, few works exist which have benchmarked the performance of the deep learning models with respect to the state-of-the-art machine learning models and prognostic scoring systems on publicly available healthcare datasets. In this paper, we present the benchmarking results for several clinical prediction tasks such as mortality prediction, length of stay prediction, and ICD-9 code group prediction using Deep Learning models, ensemble of machine learning models (Super Learner algorithm), SAPS II and SOFA scores. We used the Medical Information Mart for Intensive Care III (MIMIC-III) (v1.4) publicly available dataset, which includes all patients admitted to an ICU at the Beth Israel Deaconess Medical Center from 2001 to 2012, for the benchmarking tasks. Our results show that deep learning models consistently outperform all the other approaches especially when the 'raw' clinical time series data is used as input features to the models. Copyright © 2018 Elsevier Inc. All rights reserved.

Ion current prediction model considering columnar recombination in alpha radioactivity measurement using ionized air transportation

International Nuclear Information System (INIS)

Naito, Susumu; Hirata, Yosuke; Izumi, Mikio; Sano, Akira; Miyamoto, Yasuaki; Aoyama, Yoshio; Yamaguchi, Hiromi

2007-01-01

We present a reinforced ion current prediction model in alpha radioactivity measurement using ionized air transportation. Although our previous model explained the qualitative trend of the measured ion current values, the absolute values of the theoretical curves were about two times as large as the measured values. In order to accurately predict the measured values, we reinforced our model by considering columnar recombination and turbulent diffusion, which affects columnar recombination. Our new model explained the considerable ion loss in the early stage of ion diffusion and narrowed the gap between the theoretical and measured values. The model also predicted suppression of ion loss due to columnar recombination by spraying a high-speed air flow near a contaminated surface. This suppression was experimentally investigated and confirmed. In conclusion, we quantitatively clarified the theoretical relation between alpha radioactivity and ion current in laminar flow and turbulent pipe flow. (author)

Combining quantitative trait loci analysis with physiological models to predict genotype-specific transpiration rates.

Science.gov (United States)

Reuning, Gretchen A; Bauerle, William L; Mullen, Jack L; McKay, John K

2015-04-01

Transpiration is controlled by evaporative demand and stomatal conductance (gs ), and there can be substantial genetic variation in gs . A key parameter in empirical models of transpiration is minimum stomatal conductance (g0 ), a trait that can be measured and has a large effect on gs and transpiration. In Arabidopsis thaliana, g0 exhibits both environmental and genetic variation, and quantitative trait loci (QTL) have been mapped. We used this information to create a genetically parameterized empirical model to predict transpiration of genotypes. For the parental lines, this worked well. However, in a recombinant inbred population, the predictions proved less accurate. When based only upon their genotype at a single g0 QTL, genotypes were less distinct than our model predicted. Follow-up experiments indicated that both genotype by environment interaction and a polygenic inheritance complicate the application of genetic effects into physiological models. The use of ecophysiological or 'crop' models for predicting transpiration of novel genetic lines will benefit from incorporating further knowledge of the genetic control and degree of independence of core traits/parameters underlying gs variation. © 2014 John Wiley & Sons Ltd.

A Subject-Specific Kinematic Model to Predict Human Motion in Exoskeleton-Assisted Gait

Science.gov (United States)

Torricelli, Diego; Cortés, Camilo; Lete, Nerea; Bertelsen, Álvaro; Gonzalez-Vargas, Jose E.; del-Ama, Antonio J.; Dimbwadyo, Iris; Moreno, Juan C.; Florez, Julian; Pons, Jose L.

2018-01-01

The relative motion between human and exoskeleton is a crucial factor that has remarkable consequences on the efficiency, reliability and safety of human-robot interaction. Unfortunately, its quantitative assessment has been largely overlooked in the literature. Here, we present a methodology that allows predicting the motion of the human joints from the knowledge of the angular motion of the exoskeleton frame. Our method combines a subject-specific skeletal model with a kinematic model of a lower limb exoskeleton (H2, Technaid), imposing specific kinematic constraints between them. To calibrate the model and validate its ability to predict the relative motion in a subject-specific way, we performed experiments on seven healthy subjects during treadmill walking tasks. We demonstrate a prediction accuracy lower than 3.5° globally, and around 1.5° at the hip level, which represent an improvement up to 66% compared to the traditional approach assuming no relative motion between the user and the exoskeleton. PMID:29755336

A Subject-Specific Kinematic Model to Predict Human Motion in Exoskeleton-Assisted Gait.

Science.gov (United States)

Torricelli, Diego; Cortés, Camilo; Lete, Nerea; Bertelsen, Álvaro; Gonzalez-Vargas, Jose E; Del-Ama, Antonio J; Dimbwadyo, Iris; Moreno, Juan C; Florez, Julian; Pons, Jose L

2018-01-01

The relative motion between human and exoskeleton is a crucial factor that has remarkable consequences on the efficiency, reliability and safety of human-robot interaction. Unfortunately, its quantitative assessment has been largely overlooked in the literature. Here, we present a methodology that allows predicting the motion of the human joints from the knowledge of the angular motion of the exoskeleton frame. Our method combines a subject-specific skeletal model with a kinematic model of a lower limb exoskeleton (H2, Technaid), imposing specific kinematic constraints between them. To calibrate the model and validate its ability to predict the relative motion in a subject-specific way, we performed experiments on seven healthy subjects during treadmill walking tasks. We demonstrate a prediction accuracy lower than 3.5° globally, and around 1.5° at the hip level, which represent an improvement up to 66% compared to the traditional approach assuming no relative motion between the user and the exoskeleton.

Development of Prediction Model and Experimental Validation in Predicting the Curcumin Content of Turmeric (Curcuma longa L.).

Science.gov (United States)

Akbar, Abdul; Kuanar, Ananya; Joshi, Raj K; Sandeep, I S; Mohanty, Sujata; Naik, Pradeep K; Mishra, Antaryami; Nayak, Sanghamitra

2016-01-01

The drug yielding potential of turmeric ( Curcuma longa L.) is largely due to the presence of phyto-constituent 'curcumin.' Curcumin has been found to possess a myriad of therapeutic activities ranging from anti-inflammatory to neuroprotective. Lack of requisite high curcumin containing genotypes and variation in the curcumin content of turmeric at different agro climatic regions are the major stumbling blocks in commercial production of turmeric. Curcumin content of turmeric is greatly influenced by environmental factors. Hence, a prediction model based on artificial neural network (ANN) was developed to map genome environment interaction basing on curcumin content, soli and climatic factors from different agroclimatic regions for prediction of maximum curcumin content at various sites to facilitate the selection of suitable region for commercial cultivation of turmeric. The ANN model was developed and tested using a data set of 119 generated by collecting samples from 8 different agroclimatic regions of Odisha. The curcumin content from these samples was measured that varied from 7.2% to 0.4%. The ANN model was trained with 11 parameters of soil and climatic factors as input and curcumin content as output. The results showed that feed-forward ANN model with 8 nodes (MLFN-8) was the most suitable one with R 2 value of 0.91. Sensitivity analysis revealed that minimum relative humidity, altitude, soil nitrogen content and soil pH had greater effect on curcumin content. This ANN model has shown proven efficiency for predicting and optimizing the curcumin content at a specific site.

Development of prediction model and experimental validation in predicting the curcumin content of turmeric (Curcuma longa L.

Directory of Open Access Journals (Sweden)

Abdul Akbar

2016-10-01

Full Text Available The drug yielding potential of turmeric (Curcuma longa L. is largely due to the presence of phyto-constituent ‘curcumin’. Curcumin has been found to possess a myriad of therapeutic activities ranging from anti-inflammatory to neuroprotective. Lack of requisite high curcumin containing genotypes and variation in the curcumin content of turmeric at different agro climatic regions are the major stumbling blocks in commercial production of turmeric. Curcumin content of turmeric is greatly influenced by environmental factors. Hence, a prediction model based on artificial neural network (ANN was developed to map genome environment interaction basing on curcumin content, soli and climatic factors from different agroclimatic regions for prediction of maximum curcumin content at various sites to facilitate the selection of suitable region for commercial cultivation of turmeric. The ANN model was developed and tested using a data set of 119 generated by collecting samples from 8 different agroclimatic regions of Odisha. The curcumin content from these samples was measured that varied from 7.2% to 0.4%. The ANN model was trained with 11 parameters of soil and climatic factors as input and curcumin content as output. The results showed that feed-forward ANN model with 8 nodes (MLFN-8 was the most suitable one with R2 value of 0.91. Sensitivity analysis revealed that minimum relative humidity, altitude, soil nitrogen content and soil pH had greater effect on curcumin content. This ANN model has shown proven efficiency for predicting and optimizing the curcumin content at a specific site.

Climate Modeling and Causal Identification for Sea Ice Predictability

Energy Technology Data Exchange (ETDEWEB)

Hunke, Elizabeth Clare [Los Alamos National Lab. (LANL), Los Alamos, NM (United States); Urrego Blanco, Jorge Rolando [Los Alamos National Lab. (LANL), Los Alamos, NM (United States); Urban, Nathan Mark [Los Alamos National Lab. (LANL), Los Alamos, NM (United States)

2018-02-12

This project aims to better understand causes of ongoing changes in the Arctic climate system, particularly as decreasing sea ice trends have been observed in recent decades and are expected to continue in the future. As part of the Sea Ice Prediction Network, a multi-agency effort to improve sea ice prediction products on seasonal-to-interannual time scales, our team is studying sensitivity of sea ice to a collection of physical process and feedback mechanism in the coupled climate system. During 2017 we completed a set of climate model simulations using the fully coupled ACME-HiLAT model. The simulations consisted of experiments in which cloud, sea ice, and air-ocean turbulent exchange parameters previously identified as important for driving output uncertainty in climate models were perturbed to account for parameter uncertainty in simulated climate variables. We conducted a sensitivity study to these parameters, which built upon a previous study we made for standalone simulations (Urrego-Blanco et al., 2016, 2017). Using the results from the ensemble of coupled simulations, we are examining robust relationships between climate variables that emerge across the experiments. We are also using causal discovery techniques to identify interaction pathways among climate variables which can help identify physical mechanisms and provide guidance in predictability studies. This work further builds on and leverages the large ensemble of standalone sea ice simulations produced in our previous w14_seaice project.

A national-scale model of linear features improves predictions of farmland biodiversity.

Science.gov (United States)

Sullivan, Martin J P; Pearce-Higgins, James W; Newson, Stuart E; Scholefield, Paul; Brereton, Tom; Oliver, Tom H

2017-12-01

Modelling species distribution and abundance is important for many conservation applications, but it is typically performed using relatively coarse-scale environmental variables such as the area of broad land-cover types. Fine-scale environmental data capturing the most biologically relevant variables have the potential to improve these models. For example, field studies have demonstrated the importance of linear features, such as hedgerows, for multiple taxa, but the absence of large-scale datasets of their extent prevents their inclusion in large-scale modelling studies.We assessed whether a novel spatial dataset mapping linear and woody-linear features across the UK improves the performance of abundance models of 18 bird and 24 butterfly species across 3723 and 1547 UK monitoring sites, respectively.Although improvements in explanatory power were small, the inclusion of linear features data significantly improved model predictive performance for many species. For some species, the importance of linear features depended on landscape context, with greater importance in agricultural areas. Synthesis and applications . This study demonstrates that a national-scale model of the extent and distribution of linear features improves predictions of farmland biodiversity. The ability to model spatial variability in the role of linear features such as hedgerows will be important in targeting agri-environment schemes to maximally deliver biodiversity benefits. Although this study focuses on farmland, data on the extent of different linear features are likely to improve species distribution and abundance models in a wide range of systems and also can potentially be used to assess habitat connectivity.