Handbook of HydroInformatics: Volume I: Classic Soft-Computing Techniques

Description

Classic Soft-Computing Techniques is the first volume of the three, in the Handbook of HydroInformatics series. Through this comprehensive, 34-chapters work, the contributors explore the difference between traditional computing, also known as hard computing, and soft computing, which is based on the importance given to issues like precision, certainty and rigor. The chapters go on to define fundamentally classic soft-computing techniques such as Artificial Neural Network, Fuzzy Logic, Genetic Algorithm, Supporting Vector Machine, Ant-Colony Based Simulation, Bat Algorithm, Decision Tree Algorithm, Firefly Algorithm, Fish Habitat Analysis, Game Theory, Hybrid Cuckoo–Harmony Search Algorithm, Honey-Bee Mating Optimization, Imperialist Competitive Algorithm, Relevance Vector Machine, etc.It is a fully comprehensive handbook providing all the information needed around classic soft-computing techniques. This volume is a true interdisciplinary work, and the audience includes postgraduates and early career researchers interested in Computer Science, Mathematical Science, Applied Science, Earth and Geoscience, Geography, Civil Engineering, Engineering, Water Science, Atmospheric Science, Social Science, Environment Science, Natural Resources, and Chemical Engineering. Front Cover Handbook of HydroInformatics: Volume I: Classic Soft-Computing Techniques Copyright Dedication Contents Contributors About the editors Preface Chapter 1: Advanced machine learning techniques: Multivariate regression 1. Introduction 2. Linear regression 3. Multivariate linear regression 4. Gradient descent method 5. Polynomial regression 6. Overfitting and underfitting 7. Cross-validation 8. Comparison between linear and polynomial regressions 9. Learning curve 10. Regularized linear models 11. The ridge regression 12. The effect of collinearity in the coefficients of an estimator 13. Outliers impact 14. Lasso regression 15. Elastic net 16. Early stopping 17. Logistic regression 18. Estimation of probabilities 19. Training and the cost function 20. Conclusions Appendix: Python code Linear regression Gradient descent method Comparison between linear and polynomial regressions Learning curve The effect of collinearity in the coefficients of an estimator Outliers impact Lasso regression Elastic net Training and the cost function References Chapter 2: Bat algorithm optimized extreme learning machine: A new modeling strategy for predicting river water turbidity ... 1. Introduction 2. Study area and data 3. Methodology 3.1. Feedforward artificial neural network 3.2. Dynamic evolving neural-fuzzy inference system 3.3. Bat algorithm optimized extreme learning machine 3.4. Multiple linear regression 3.5. Performance assessment of the models 4. Results and discussion 4.1. USGS 1497500 station 4.2. USGS 11501000 station 4.3. USGS 14210000 station 4.4. USGS 14211010 station 5. Conclusions References Chapter 3: Bayesian theory: Methods and applications 1. Introduction 2. Bayesian inference 3. Phases 4. Estimates 5. Theorem Bayes 5.1. Argument of Bayes 5.2. Bayesian estimation theory 5.3. Machine learning using Bayesian method 5.4. Bayesian theory in machine learning 5.5. Definition of basic concepts 5.6. Bayesian machine learning methods 5.7. Optimal Bayes classifier 5.7.1. Background and theory 5.8. Naive Bayes classifier 6. Bayesian network 7. History of Bayesian model application in water resources 8. Case study of Bayesian network application in modeling of evapotranspiration of reference plant 9. Conclusions References Chapter 4: CFD models 1. Introduction 2. Numerical model of one-dimensional advection dispersion equation (1D-ADE) 3. Physically influenced scheme 4. Finite Volume Solution of Saint-Venant equations for dam-break simulation using PIS 5. Discretization of continuity equation using PIS 6. Discretization of the momentum equation using PIS 7. Quasi-two-dimensional flow simulation 8. Numerical solution of quasi-two-dimensional model 9. 3D numerical modeling of flow in compound channel using turbulence models 10. Three-dimensional numerical model 11. Grid generation and the flow filed solution 12. Comparison of different turbulence models 13. Three-dimensional pollutant transfer modeling 14. Results of pollutant transfer modeling 15. Conclusions References Chapter 5: Cross-validation 1. Introduction 1.1. Importance of validation 1.2. Validation of the training process 2. Cross-validation 2.1. Exhaustive and nonexhaustive cross-validation 2.2. Repeated random subsampling cross-validation 2.3. Time-series cross-validation 2.4. k-fold cross-validation 2.5. Stratified k-fold cross-validation 2.6. Nested 3. Computational procedures 4. Conclusions References Chapter 6: Comparative study on the selected node and link-based performance indices to investigate the hydraulic capacit ... 1. Introduction 2. Resilience of water distribution network 3. Hydraulic uniformity index (HUI) 4. Mean excess pressure (MEP) 5. Proposed measure 5.1. Energy loss uniformity (ELU) 6. Hanoi network 7. Results and discussion 8. Conclusions References Chapter 7: The co-nodal system analysis 1. Introduction 2. Co-nodal and system analysis 3. Paleo-hydrology and remote sensing 4. Methods 5. Nodes and cyclic confluent system 5.1. H-cycloids analysis and fluvial dynamics 6. Three Danube phases 7. Danubian hypocycles as overlapping phases 8. Conclusions References Further reading Chapter 8: Data assimilation 1. Introduction 2. What is data assimilation? 3. Types of data assimilation methods 3.1. Types of updating procedure 3.1.1. Variational data assimilation 3.1.2. Sequential data assimilation 3.2. Types of updating variable 3.2.1. Updating input variable 3.2.2. Updating model parameter 3.2.3. Updating state variable 3.2.4. Updating output variable 4. Optimal filtering methods 4.1. Kalman filter 4.1.1. Kalman filter limitations 4.2. Transfer function 4.3. Extended Kalman filter 4.4. Unscented Kalman filter 5. Auto-regressive method 6. Considerations in using data assimilation 7. Conclusions References Chapter 9: Data reduction techniques 1. Introduction 2. Principal component analysis 3. Singular spectrum analysis 3.1. Univariate singular spectral analysis 3.2. Multivariate singular spectral analysis 4. Canonical correlation analysis 5. Factor analysis 5.1. Principal axis factoring 6. Random projection 7. Isometric mapping 8. Self-organizing maps 9. Discriminant analysis 10. Piecewise aggregate approximation 11. Clustering 11.1. k-means clustering 11.2. Hierarchical clustering 11.3. Density-based clustering 12. Conclusions References Chapter 10: Decision tree algorithms 1. Introduction 1.1. ID3 algorithm 1.2. C4.5 algorithm 1.3. CART algorithm 1.4. CHAID algorithm 1.5. M5 algorithm 1.6. Random forest 1.7. Application of DT algorithms in water sciences 2. M5 model tree 2.1. Splitting 2.2. Pruning 2.3. Smoothing 3. Data set 3.1. Empirical formula for flow discharge 3.2. Model evaluation and comparison 4. Modeling and results 4.1. Initial tree 4.2. Pruning 4.3. Comparing M5 model and empirical formula 5. Conclusions References Chapter 11: Entropy and resilience indices 1. Introduction 2. Water resource and infrastructure performance evaluation 3. Entropy 3.1. Thermodynamic entropy 3.2. Statistical-mechanical entropy 3.3. Information entropy 3.4. Application of entropy in water resources area 4. Resilience 4.1. Application of resilience in water resources area 4.2. Resilience in UWS 4.3. Resilience in urban environments 4.4. Resilience to floods 4.5. Resilience to drought 5. Conclusions References Chapter 12: Forecasting volatility in the stock market data using GARCH, EGARCH, and GJR models 1. Introduction 2. Methodology 2.1. Types of GARCH models 2.1.1. GARCH model 2.1.2. EGARCH model 2.1.3. GJR model 3. Application and results 4. Conclusions References Chapter 13: Gene expression models 1. Introduction 2. Genetic programming 2.1. The basic steps in GEP development 2.2. The basic steps in GEP development 3. Tree-based GEP 3.1. Tree depth control 3.2. Maximum tree depth 3.3. Penalizing the large trees 3.4. Dynamic maximum-depth technique 4. Linear genetic programming 5. Evolutionary polynomial regression 6. Multigene genetic programming 7. Pareto optimal-multigene genetic programming 8. Some applications of GEP-based models in hydro informatics 8.1. Derivation of quadric polynomial function using GEP 8.2. Derivation of Colebrook-White equation using GEP 8.3. Derivation of the exact form of shields diagram using GEP 8.4. Extraction of regime river equations using GEP 8.5. Extraction of longitudinal dispersion coefficient equations using GEP 9. Conclusions References Chapter 14: Gradient-based optimization 1. Introduction 2. Materials and method 2.1. GRG solver 3. Results and discussion 3.1. Solving nonlinear equations 3.2. Application in parameter estimation 3.3. Fitting empirical equations 4. Conclusions References Chapter 15: Gray wolf optimization algorithm 1. Introduction 2. Theory of GWO 3. Mathematical modeling of gray wolf optimizer 3.1. Social hierarchy 3.2. Encircling prey 3.3. Hunting behavior 3.4. Exploitation in GWO-attacking prey 3.5. Exploration in GWO-search for prey 4. Gray wolf optimization example for reservoir operation 5. Conclusions Appendix A: GWO Matlab codes for the reservoir example References Chapter 16: Kernel-based modeling 1. Introduction 2. Support vector machine 2.1. Support vector classification 2.1.1. Linear classifiers 2.1.2. Non-linear classifiers and kernels application 2.2. Support vector regression 3. Gaussian processes 3.1. Gaussian process regression 3.2. Gaussian process classification 4. Kernel extreme learning machine 5. Kernels type 5.1. Fisher kernel 5.2. Graph kernels 5.3. Kernel smoother 5.4. Polynomial kernel 5.5. Radial basis function kernel 5.6. Pearson kernel 5.7. String kernels 5.8. Neural tangent kernel 6. Application of kernel-based approaches 6.1. Total resistance and form resistance of movable bed channels 6.2. Energy losses of rectangular and circular culverts 6.3. Lake and reservoir water level prediction 6.4. Streamflow forecasting 6.5. Sediment load prediction 6.6. Pier scour modeling 6.7. Reservoir evaporation prediction 7. Conclusions References Further reading Chapter 17: Large eddy simulation: Subgrid-scale modeling with neural network 1. Introduction 2. LES and traditional subgrid-scale modeling 3. Data-driven LES closures 4. Guidelines for SGS modeling 4.1. Simulation project definition 4.2. A priory analysis with DNS 4.3. Neural network based SGS model construction 5. Conclusions References Chapter 18: Lattice Boltzmann method and its applications 1. Introduction 2. Lattice Boltzmann equations 2.1. BGK approximation 2.2. Lattice Boltzmann models 2.3. Multirelaxation time lattice Boltzmann (MRT) 2.4. Boundary conditions 2.4.1. Bounce back 2.4.2. The boundary with a given velocity 2.4.3. The boundary with given pressure 2.4.4. Open boundary condition 2.4.5. Symmetry boundary condition 2.4.6. Periodic boundary condition 3. Thermal LBM 3.1. Boundary condition with a given temperature 3.2. Constant heat flux boundary condition 4. Multicomponent LBM (species transport modeling) 5. Flow simulation in porous media 6. Dimensionless numbers 7. Flow chart of the simulation procedure 8. Multiphase flows 8.1. The color-gradient model 8.2. Shan-Chen model 9. Sample test cases and codes 9.1. Free convection in L-cavity 9.2. Force convection in a channel 10. Conclusions Appendix A Computer code for free convection in L-cavity Appendix B Computer code for force convection in a channel References Chapter 19: Multigene genetic programming and its various applications 1. Introduction 2. Genetic programming and its variants 3. An introduction to multigene genetic programming 4. Main controlling parameters of MGGP 5. A review on MGGP applications 6. Future trends of MGGP applications 7. A case study of the MGGP application 8. Conclusions References Chapter 20: Ontology-based knowledge management framework in business organizations and water users networks in Tanzania 1. Introduction 2. Theoretical framework 3. Empirical literature 4. Ontology-based knowledge management framework in business organizations: A conceptual framework 5. Ontology-based knowledge management framework in business organizations and water user networks proposed system 6. The practice of knowledge organization and expression 6.1. Ontology 6.2. Knowledge representation and organization base on ontology 6.3. Knowledge retrieval base ontology 6.4. Knowledge application and implementation base on ontology 7. Conclusions References Chapter 21: Parallel chaos search-based incremental extreme learning machine 1. Introduction 2. Materials and methods 2.1. Study area description 2.2. Modeling approaches 2.2.1. Parallel chaos search based incremental extreme learning machine 2.3. Performance assessment of the models 3. Results and discussion 4. Conclusions References Chapter 22: Relevance vector machine (RVM) 1. Introduction 2. Machine learning algorithms 2.1. Supervised learning 2.2. Unsupervised learning 3. Support vector machine 4. Relevance vector machine 4.1. Measurement model representation 4.2. Relevance vector regression 4.3. Relevance vector classification 4.4. Limitations and performance analysis 4.5. Multivariate relevance vector machines 5. Preprocessing step 5.1. Data normalization 5.2. Data reduction 5.3. Dataset split ratio 5.3.1. Holdout method 5.3.2. Random subsampling 5.3.3. Cross-validation 5.3.4. Leaving-one-out 6. Applications of relevance vector machine 6.1. Sediment concentration estimation 6.2. Drought monitoring 6.3. Groundwater quality monitoring 6.4. Evaporative losses in reservoirs 6.5. Environmental science 7. Conclusions References Chapter 23: Stochastic learning algorithms 1. Introduction 2. Gradient descent 2.1. Theory of batch gradient descent 2.2. Theory of SGD 3. Perceptron 3.1. Theory of perceptron 3.2. Perceptron learning procedure 4. Adaline 4.1. Theory of Adaline 4.2. Adaline learning procedure 5. Multilayer network 5.1. Multilayer network learning procedure 6. Learning vector quantization 6.1. LVQ learning procedure 7. K-means clustering 7.1. What is clustering? 7.2. Theory of K-means 8. Gradient boosting 8.1. What is boosting? 8.2. Theory of gradient boosting (GB) 8.3. Stochastic gradient boosting 9. Conclusions References Appendix A Appendix B Appendix C Appendix D Appendix E Chapter 24: Supporting vector machines 1. Introduction 2. SVMs for classification problems 2.1. Linear classifiers 2.2. Non-linear classifiers 3. SVMs for regression problems 4. Selection of SVM parameters 4.1. Margin 4.2. Regularization 4.3. Kernels 4.4. Gamma parameter 5. Application of support vector machines 5.1. Application of support vector regression in the water recourse engineering 5.1.1. Some examples from literature reviews for investigating hydraulic problems 5.1.2. Some examples from literature reviews for investigating hydrological problems 5.1.3. Some examples from literature reviews for investigating water resource management problems 6. Conclusions References Chapter 25: Uncertainty analysis using fuzzy models in hydroinformatics 1. Introduction 2. Fuzzy logic theory 2.1. Fuzzification 2.2. Rule base 2.3. Inference 2.4. Defuzzification 3. Concept of fuzzy uncertainty analysis 4. Uncertainty analysis applications 4.1. Flood forecasting 4.2. Groundwater modeling 5. Machine learning and fuzzy sets 6. Fuzzy sets and probabilistic approach 7. Conclusions References Chapter 26: Uncertainty-based resiliency evaluation 1. Introduction 2. Uncertainty analysis by the first-order method 3. Risk and resilience analysis 4. Reliability computation by direct integration 5. Reliability computation using safety margin/safety factor 6. Safety margin 7. Safety factor 8. Uncertainty-based hydraulic designs 9. Hydrologic uncertainties 10. Hydraulics uncertainties 11. Monte-Carlo uncertainty analysis in quasi-2D model parameters 12. SKM model 13. Uncertainty based river flow modeling with Monte-Carlo simulator 14. Monte-Carlo uncertainty analysis in machine learning techniques 15. Uncertainty evaluation using the integrated Bayesian multimodel framework 16. Copula-based uncertainty analysis 17. Uncertainty analysis with Tsallis entropy 18. Theory of evidence for uncertainty in hydroinformatics 19. Resiliency quantification 20. Conclusions References Index Back Cover