Refining the Committee Approach and Uncertainty Prediction in Hydrological Modelling

Wysyłka: Niedostępna
Sugerowana cena detaliczna 291,00 PLN
Nasza cena: 272,09 PLN
Oszczędzasz 6%
Dodaj do Schowka
Zaloguj się
Przypomnij hasło
×
×
Paypo
Oferujemy szeroki asortyment - ponad 120 tys. produktów
Dysponujemy solidną wiedzą - działamy już 11 lat
Dbamy o wybór najcenniejszych tytułów

Opis: Refining the Committee Approach and Uncertainty Prediction in Hydrological Modelling - Nagendra Kayastha

Due to the complexity of hydrological systems a single model may be unable to capture the full range of a catchment response and accurately predict the streamflows. A solution could be the in use of several specialized models organized in the so-called committees. Refining the committee approach is one of the important topics of this study, and it is demonstrated that it allows for increased predictive capability of models. Another topic addressed is the prediction of hydrologic models' uncertainty. The traditionally used Monte Carlo method is based on the past data and cannot be directly used for estimation of model uncertainty for the future model runs during its operation. In this thesis the so-called MLUE (Machine Learning for Uncertainty Estimation) approach is further explored and extended; in it the machine learning techniques (e.g. neural networks) are used to encapsulate the results of Monte Carlo experiments in a predictive model that is able to estimate uncertainty for the future states of the modelled system. Furthermore, it is demonstrated that a committee of several predictive uncertainty models allows for an increase in prediction accuracy. Catchments in Nepal, UK and USA are used as case studies. In flood modelling hydrological models are typically used in combination with hydraulic models forming a cascade, often supported by geospatial processing. For uncertainty analysis of flood inundation modelling of the Nzoia catchment (Kenya) SWAT hydrological and SOBEK hydrodynamic models are integrated, and the parametric uncertainty of the hydrological model is allowed to propagate through the model cascade using Monte Carlo simulations, leading to the generation of the probabilistic flood maps. Due to the high computational complexity of these experiments, the high performance (cluster) computing framework is designed and used. This study refined a number of hydroinformatics techniques, thus enhancing uncertainty-based hydrological and integrated modelling.SUMMARY 1 INTRODUCTION 1.1 Background 1.1.1 Conceptual hydrological models 1.1.2 Committee hydrological models (multi-models) 1.1.3 Uncertainty analysis of hydrological models 1.1.4 Uncertainty analysis using machine learning techniques 1.1.5 Committee of predictive uncertainty models 1.1.6 Flood inundation models and their uncertainty 1.2 Research questions 1.3 Research objectives 1.4 Case studies 1.4.1 Alzette catchment 1.4.2 Bagmati catchment 1.4.3 Brue catchment 1.4.4 Leaf catchment 1.4.5 Nzoia catchment 1.5 Terminology 1.6 Outline of the thesis 2 CONCEPTUAL AND DATA-DRIVEN HYDROLOGICAL MODELLING 2.1 Introduction 2.2 HBV hydrological models for the considered case studies 2.2.1 HBV model brief characterization 2.2.2 Software development of HBV model 2.2.3 Models setup 2.2.3.1. HBV model setup for the Brue catchment 2.2.3.2. HBV model setup for the Bagmati catchment 2.2.3.3. HBV model setup for the Nzoia catchment 2.2.3.4. HBV model setup for the Leaf catchment 2.2.3.5. HBV model setup for the Alzette catchment 2.3 SWAT model for the Nzoia catchment 2.3.1 SWAT model description 2.3.2 Inputs for the SWAT model 2.4 Calibration of hydrological models 2.4.1 Single objective optimization 2.4.2 Multi objective optimization 2.4.3 SWAT-NSGAX tool and its application 2.5 Data driven modelling 2.5.1 Introduction 2.5.2 Machine learning in data-driven rainfall-runoff modelling 2.5.3 Artificial neural networks 2.5.4 Model trees 2.5.5 Locally weighted regression 2.5.6 Selection of input variables 2.5.7 Data-driven rainfall-runoff model of the Bagmati catchment 2.5.8 Data-driven rainfall-runoff model of the Leaf catchment 2.6 Summary 3 COMMITTEES OF HYDROLOGICAL MODELS 3.1 Introduction 3.2 Specialized hydrological models 3.3 Committees of specialized models 3.3.1 Fuzzy committee models 3.3.2 States-based committee models 3.3.3 Inputs-based committee models 3.3.4 Outputs-based committee models 3.4 Performance measures 3.5 Models setup for Alzette, Bagmati and Leaf catchments 3.6 Results and discussion 3.6.1 Fuzzy committee models 3.6.2 States-, inputs-, and outputs-based committee models for Brue, Bagmati, and Leaf 3.7 Summary 4 HYBRID COMMITTEES OF HYDROLOGICAL MODELS 4.1 Introduction 4.2 Low flows simulation 4.3 ANN models specialized on low flows 4.4 Committee of ANN and HBV for Bagmati and Leaf 4.5 Results and discussion 4.6 Summary 5 MODEL PARAMETRIC UNCERTAINTY AND EFFECTS OF SAMPLING STRATEGIES 5.1 Introduction 5.2 Comparison of parameter estimation and uncertainty analysis methods 5.3 Sampling strategies for uncertainty analysis of hydrological model 5.3.1 Monte Carlo simulation 5.3.2 GLUE 5.3.3 MCMC 5.3.4 SCEMUA 5.3.5 DREAM 5.3.6 ACCO 5.3.7 PSO 5.4 Characterization of uncertainty 5.4.1 Prediction interval 5.4.2 Uncertainty indices 5.4.3 Likelihood functions 5.4.3.1. Informal likelihood 5.4.3.2. Formal Bayesian likelihood 5.5 Experiment setup for the Nzoia catchment 5.6 Experimental results and discussion 5.6.1 Distribution of the model objective function 5.6.2 Parameter posterior distribution 5.6.3 Statistical analysis of results 5.7 Summary 6 PREDICTION OF UNCERTAINTY BY MACHINE LEARNING TECHNIQUES 6.1 Introduction 6.2 Machine learning techniques for building predictive uncertainty models 6.2.1 Characterization of uncertainty 6.2.2 Techniques for building predictive uncertainty models 6.2.3 Selection of input variables for the predictive uncertainty model 6.2.4 Verification of the predictive uncertainty models 6.3 Experimental setup 6.3.1 Uncertainty analysis for case studies Bagmati and Brue 6.3.2 Machine learning models (ANN, MT and LWR) 6.3.3 Modelling the probability distribution function 6.4 Results and discussion 6.4.1 Comparison among ANN, MT and LWR 6.5 Summary 7 COMMITTEES OF MODELS PREDICTING MODELS' UNCERTAINTY 7.1 Introduction 7.2 Bayesian Model Averaging 7.3 Building predictive uncertainty models for the Bagmati catchment 7.3.1 Several sets of variables 7.3.2 Model averaging results and discussion 7.4 Building predictive uncertainty models for the Nzoia catchment 7.4.1 Committee of uncertainty prediction models 7.4.2 Results and discussion 7.5 Summary 8 INTEGRATION OF HYDROLOGICAL AND HYDRODYNAMIC MODELS AND THEIR UNCERTAINTY IN INUNDATION MODELLING 8.1 Introduction 8.2 Flood models 8.3 Model integration 8.4 Propagation of uncertainties in integrated models 8.5 SWAT and SOBEK models setup for the Nzoia catchment 8.6 Approach to estimate the uncertainty of flood inundation extent 8.7 Use of parallel computing 8.8 Quantification of model performance and uncertainty 8.9 Results and discussion 8.10 Summary 9 CONCLUSIONS AND RECOMMENDATIONS 9.1 Committee modelling 9.2 Sampling-based uncertainty analysis techniques 9.3 Uncertainty prediction using machine learning techniques 9.4 Committee of predictive uncertainty models 9.5 Uncertainty analysis of flood inundation models 9.6 Final conclusion REFERENCES LIST OF ACRONYMS LIST OF TABLES LIST OF FIGURES SAMENVATTING ACKNOWLEDGEMENT ABOUT THE AUTHOR


Szczegóły: Refining the Committee Approach and Uncertainty Prediction in Hydrological Modelling - Nagendra Kayastha

Tytuł: Refining the Committee Approach and Uncertainty Prediction in Hydrological Modelling
Autor: Nagendra Kayastha
Producent: Routledge
ISBN: 9781138027466
Rok produkcji: 2015
Ilość stron: 200
Oprawa: Miękka
Waga: 0.36 kg


Recenzje: Refining the Committee Approach and Uncertainty Prediction in Hydrological Modelling - Nagendra Kayastha

Zaloguj się
Przypomnij hasło
×
×

Refining the Committee Approach and Uncertainty Prediction in Hydrological Modelling

Due to the complexity of hydrological systems a single model may be unable to capture the full range of a catchment response and accurately predict the streamflows. A solution could be the in use of several specialized models organized in the so-called committees. Refining the committee approach is one of the important topics of this study, and it is demonstrated that it allows for increased predictive capability of models. Another topic addressed is the prediction of hydrologic models' uncertainty. The traditionally used Monte Carlo method is based on the past data and cannot be directly used for estimation of model uncertainty for the future model runs during its operation. In this thesis the so-called MLUE (Machine Learning for Uncertainty Estimation) approach is further explored and extended; in it the machine learning techniques (e.g. neural networks) are used to encapsulate the results of Monte Carlo experiments in a predictive model that is able to estimate uncertainty for the future states of the modelled system. Furthermore, it is demonstrated that a committee of several predictive uncertainty models allows for an increase in prediction accuracy. Catchments in Nepal, UK and USA are used as case studies. In flood modelling hydrological models are typically used in combination with hydraulic models forming a cascade, often supported by geospatial processing. For uncertainty analysis of flood inundation modelling of the Nzoia catchment (Kenya) SWAT hydrological and SOBEK hydrodynamic models are integrated, and the parametric uncertainty of the hydrological model is allowed to propagate through the model cascade using Monte Carlo simulations, leading to the generation of the probabilistic flood maps. Due to the high computational complexity of these experiments, the high performance (cluster) computing framework is designed and used. This study refined a number of hydroinformatics techniques, thus enhancing uncertainty-based hydrological and integrated modelling.SUMMARY 1 INTRODUCTION 1.1 Background 1.1.1 Conceptual hydrological models 1.1.2 Committee hydrological models (multi-models) 1.1.3 Uncertainty analysis of hydrological models 1.1.4 Uncertainty analysis using machine learning techniques 1.1.5 Committee of predictive uncertainty models 1.1.6 Flood inundation models and their uncertainty 1.2 Research questions 1.3 Research objectives 1.4 Case studies 1.4.1 Alzette catchment 1.4.2 Bagmati catchment 1.4.3 Brue catchment 1.4.4 Leaf catchment 1.4.5 Nzoia catchment 1.5 Terminology 1.6 Outline of the thesis 2 CONCEPTUAL AND DATA-DRIVEN HYDROLOGICAL MODELLING 2.1 Introduction 2.2 HBV hydrological models for the considered case studies 2.2.1 HBV model brief characterization 2.2.2 Software development of HBV model 2.2.3 Models setup 2.2.3.1. HBV model setup for the Brue catchment 2.2.3.2. HBV model setup for the Bagmati catchment 2.2.3.3. HBV model setup for the Nzoia catchment 2.2.3.4. HBV model setup for the Leaf catchment 2.2.3.5. HBV model setup for the Alzette catchment 2.3 SWAT model for the Nzoia catchment 2.3.1 SWAT model description 2.3.2 Inputs for the SWAT model 2.4 Calibration of hydrological models 2.4.1 Single objective optimization 2.4.2 Multi objective optimization 2.4.3 SWAT-NSGAX tool and its application 2.5 Data driven modelling 2.5.1 Introduction 2.5.2 Machine learning in data-driven rainfall-runoff modelling 2.5.3 Artificial neural networks 2.5.4 Model trees 2.5.5 Locally weighted regression 2.5.6 Selection of input variables 2.5.7 Data-driven rainfall-runoff model of the Bagmati catchment 2.5.8 Data-driven rainfall-runoff model of the Leaf catchment 2.6 Summary 3 COMMITTEES OF HYDROLOGICAL MODELS 3.1 Introduction 3.2 Specialized hydrological models 3.3 Committees of specialized models 3.3.1 Fuzzy committee models 3.3.2 States-based committee models 3.3.3 Inputs-based committee models 3.3.4 Outputs-based committee models 3.4 Performance measures 3.5 Models setup for Alzette, Bagmati and Leaf catchments 3.6 Results and discussion 3.6.1 Fuzzy committee models 3.6.2 States-, inputs-, and outputs-based committee models for Brue, Bagmati, and Leaf 3.7 Summary 4 HYBRID COMMITTEES OF HYDROLOGICAL MODELS 4.1 Introduction 4.2 Low flows simulation 4.3 ANN models specialized on low flows 4.4 Committee of ANN and HBV for Bagmati and Leaf 4.5 Results and discussion 4.6 Summary 5 MODEL PARAMETRIC UNCERTAINTY AND EFFECTS OF SAMPLING STRATEGIES 5.1 Introduction 5.2 Comparison of parameter estimation and uncertainty analysis methods 5.3 Sampling strategies for uncertainty analysis of hydrological model 5.3.1 Monte Carlo simulation 5.3.2 GLUE 5.3.3 MCMC 5.3.4 SCEMUA 5.3.5 DREAM 5.3.6 ACCO 5.3.7 PSO 5.4 Characterization of uncertainty 5.4.1 Prediction interval 5.4.2 Uncertainty indices 5.4.3 Likelihood functions 5.4.3.1. Informal likelihood 5.4.3.2. Formal Bayesian likelihood 5.5 Experiment setup for the Nzoia catchment 5.6 Experimental results and discussion 5.6.1 Distribution of the model objective function 5.6.2 Parameter posterior distribution 5.6.3 Statistical analysis of results 5.7 Summary 6 PREDICTION OF UNCERTAINTY BY MACHINE LEARNING TECHNIQUES 6.1 Introduction 6.2 Machine learning techniques for building predictive uncertainty models 6.2.1 Characterization of uncertainty 6.2.2 Techniques for building predictive uncertainty models 6.2.3 Selection of input variables for the predictive uncertainty model 6.2.4 Verification of the predictive uncertainty models 6.3 Experimental setup 6.3.1 Uncertainty analysis for case studies Bagmati and Brue 6.3.2 Machine learning models (ANN, MT and LWR) 6.3.3 Modelling the probability distribution function 6.4 Results and discussion 6.4.1 Comparison among ANN, MT and LWR 6.5 Summary 7 COMMITTEES OF MODELS PREDICTING MODELS' UNCERTAINTY 7.1 Introduction 7.2 Bayesian Model Averaging 7.3 Building predictive uncertainty models for the Bagmati catchment 7.3.1 Several sets of variables 7.3.2 Model averaging results and discussion 7.4 Building predictive uncertainty models for the Nzoia catchment 7.4.1 Committee of uncertainty prediction models 7.4.2 Results and discussion 7.5 Summary 8 INTEGRATION OF HYDROLOGICAL AND HYDRODYNAMIC MODELS AND THEIR UNCERTAINTY IN INUNDATION MODELLING 8.1 Introduction 8.2 Flood models 8.3 Model integration 8.4 Propagation of uncertainties in integrated models 8.5 SWAT and SOBEK models setup for the Nzoia catchment 8.6 Approach to estimate the uncertainty of flood inundation extent 8.7 Use of parallel computing 8.8 Quantification of model performance and uncertainty 8.9 Results and discussion 8.10 Summary 9 CONCLUSIONS AND RECOMMENDATIONS 9.1 Committee modelling 9.2 Sampling-based uncertainty analysis techniques 9.3 Uncertainty prediction using machine learning techniques 9.4 Committee of predictive uncertainty models 9.5 Uncertainty analysis of flood inundation models 9.6 Final conclusion REFERENCES LIST OF ACRONYMS LIST OF TABLES LIST OF FIGURES SAMENVATTING ACKNOWLEDGEMENT ABOUT THE AUTHOR

Powiadom o dostępności
Podaj swój e-mail a zostaniesz poinformowany jak tylko pozycja będzie dostępna.
×
Cena 291,00 PLN
Nasza cena 272,09 PLN
Oszczędzasz 6%
Wysyłka: Niedostępna
Dodaj do Schowka
Zaloguj się
Przypomnij hasło
×
×

Paypo

Szczegóły: Refining the Committee Approach and Uncertainty Prediction in Hydrological Modelling - Nagendra Kayastha

Tytuł: Refining the Committee Approach and Uncertainty Prediction in Hydrological Modelling
Autor: Nagendra Kayastha
Producent: Routledge
ISBN: 9781138027466
Rok produkcji: 2015
Ilość stron: 200
Oprawa: Miękka
Waga: 0.36 kg


Recenzje: Refining the Committee Approach and Uncertainty Prediction in Hydrological Modelling - Nagendra Kayastha

Zaloguj się
Przypomnij hasło
×
×

Klienci, którzy kupili oglądany produkt kupili także:


Zaloguj się
Przypomnij hasło
×
×
Dodane do koszyka
×