Atnaujinkite slapukų nuostatas

El. knyga: Predicting Storm Surges: Chaos, Computational Intelligence, Data Assimilation and Ensembles: UNESCO-IHE PhD Thesis [Taylor & Francis e-book]

(UNESCO-IHE Institute for Water Education, Delft, The Netherlands)
  • Formatas: 200 pages
  • Išleidimo metai: 16-Dec-2011
  • Leidėjas: CRC Press
  • ISBN-13: 9780429088261
Kitos knygos pagal šią temą:
  • Taylor & Francis e-book
  • Kaina: 73,85 €*
  • * this price gives unlimited concurrent access for unlimited time
  • Standartinė kaina: 105,50 €
  • Sutaupote 30%
Predicting Storm Surges: Chaos, Computational Intelligence, Data Assimilation and Ensembles: UNESCO-IHE PhD ThesisDidesnis paveikslėlis
  • Formatas: 200 pages
  • Išleidimo metai: 16-Dec-2011
  • Leidėjas: CRC Press
  • ISBN-13: 9780429088261
Kitos knygos pagal šią temą:
Taylor & Francis e-booksMore info about Taylor & Francis e-books
Partnerių interneto svetainė: https://www.taylorfrancis.com/books/9780429088261
In his December 2011 doctoral dissertation in hydroinformatics at the Delft University of Technology, Siek presents a new approach to modeling to predict storm surges on coastlines. The models currently in use in the Netherlands are process models--also called physically-based or numerical. He focuses by contrast on data-driven modeling, which primarily uses the analysis of the data characterizing the underlying system. The model is defined mainly on the basis of connections between system state variables, with only a limited knowledge of the details about the physical behavior of the system. His goal is to build a more accurate, chaotic model that can complement existing operational storm surge models for the North Sea region. He has not indexed his study. Annotation ©2012 Book News, Inc., Portland, OR (booknews.com)

Accurate predictions of storm surge are of importance in many coastal areas in the world to avoid and mitigate its destructive impacts. For this purpose the physically-based (process) numerical models are typically utilized. However, in data-rich cases, one may use data-driven methods aiming at reconstructing the internal patterns of the modelled processes and relationships between the observed descriptive variables. This book focuses on data-driven modelling using methods of nonlinear dynamics and chaos theory. First, some fundamentals of physical oceanography, nonlinear dynamics and chaos, computational intelligence and European operational storm surge models are covered. After that a number of improvements in building chaotic models are presented: nonlinear time series analysis, multi-step prediction, phase space dimensionality reduction, techniques dealing with incomplete time series, phase error correction, finding true neighbours, optimization of chaotic model, data assimilation and multi-model ensemble prediction. The major case study is surge prediction in the North Sea, with some tests on a Caribbean Sea case. The modelling results showed that the enhanced predictive chaotic models can serve as an efficient tool for accurate and reliable short and mid-term predictions of storm surges in order to support decision-makers for flood prediction and ship navigation.

Chapter 1 Introduction
1(14)
1.1 Motivation: Natural disasters
1(2)
1.2 Modeling Natural Phenomena: Hydroinformatics
3(2)
1.3 Predicting Storm Surges
5(4)
1.3.1 Physically-based modeling
6(1)
1.3.2 Data driven modeling: Nonlinear dynamics and chaos theory
7(1)
1.3.3 Main relations between the two modeling paradigms: chaotic modeling
8(1)
1.4 Chaotic Behaviors in Ocean Surge and Other Aquatic Phenomena
9(1)
1.5 Main Objectives
10(2)
1.6 Thesis Outline
12(3)
Chapter 2 Case Study
15(10)
2.1 Study Area: The North Sea
15(2)
2.2 North Sea Characteristics
17(2)
2.2.1 Ocean dynamics
17(1)
2.2.2 Tides and sea level
18(1)
2.3 Storm Surge Condition in the North Sea
19(3)
2.3.1 Storm Surge Warning Service
21(1)
2.3.2 Procedure for issuing warnings and alarms
21(1)
2.4 Data Description
22(1)
2.5 Summary
23(2)
Chapter 3 Storm Surge Modeling
25(22)
3.1 Introduction
25(1)
3.2 Physical Oceanography
25(6)
3.2.1 Ocean waves and its classification
25(2)
3.2.1.1 Water depth
27(1)
3.2.1.2 Method of waves generation
28(1)
3.2.1.3 Period of waves
28(1)
3.2.1.4 Relationship to the Generating Force
28(1)
3.2.2 Tides
29(2)
3.3 Surges
31(2)
3.3.1 Tide-Surge Interaction
33(1)
3.4 SWAN Wave Spectrum Model
33(2)
3.5 Physcially-based Storm Surge Prediction Model
35(1)
3.6 European Meteorological Offices and Storm Surge Models
36(7)
3.6.1 North West Shelf Operational Oceanographic System (NOOS)
36(1)
3.6.2 KNMI and RIKZ
37(5)
3.6.3 European Centre for Medium-Range Weather Predictions (ECMWF)
42(1)
3.7 Linking Predictive Chaotic Model with European Operational Storm Surge Models
43(2)
3.8 Summary
45(2)
Chapter 4 Computational Intelligence
47(20)
4.1 Introduction
47(3)
4.2 Artificial Neural Networks
50(8)
4.2.1 Mathematical model of artificial neuron
52(1)
4.2.2 Learning methods
53(2)
4.2.3 Multi-layer perceptron and back-propagation algorithm
55(2)
4.2.4 Dynamic neural network
57(1)
4.3 Instance-Based Learning
58(3)
4.3.1 k-nearest neighbors learning
59(1)
4.3.2 Distance weighted nearest neighbors algorithm
60(1)
4.3.3 Locally weighted regression
60(1)
4.4 Hierarchical Modular Models
61(3)
4.5 Evolutionary and Other Randomized Search Algorithms
64(1)
4.6 Summary
65(2)
Chapter 5 Nonlinear Dynamics and Chaos Theory
67(24)
5.1 Introduction
67(1)
5.2 Basics of Chaos
68(4)
5.2.1 Dynamical system
68(1)
5.2.2 Phase space
69(1)
5.2.3 Various behaviors of dynamical system
69(1)
5.2.4 Dynamical invariants
70(1)
5.2.5 Chaos in Iterative Maps
70(2)
5.3 Geometrical analysis of maps
72(1)
5.3.1 Cobweb method
72(1)
5.3.2 Return plot
72(1)
5.3.3 Fixed points and stability analysis
73(1)
5.4 Bifurcations
73(1)
5.5 Nonlinear Dynamics in Differential Equations
74(3)
5.5.1 Sensitivity to initial conditions
75(1)
5.5.2 Properties of chaos
76(1)
5.6 Phase Space Reconstruction -- Method of Time Delay
77(1)
5.7 Finding appropriate time delay
78(1)
5.8 Estimating embedding dimension
79(4)
5.8.1 Self-similarity: Dimension
79(2)
5.8.2 False nearest neighbors
81(1)
5.8.3 Cao's method
81(1)
5.8.4 Kolmogorov-Sinai Entropy
82(1)
5.9 Analysis of Stability: Lyapunov Exponents
83(2)
5.10 Building Chaotic Model
85(3)
5.11 Recurrence Plots
88(2)
5.12 Summary
90(1)
Chapter 6 Building Predictive Chaotic Model
91(26)
6.1 Introduction
91(2)
6.2 Power Spectral Density: Periodicity and Stochasticity
93(1)
6.3 Phase Space Reconstruction: Finding Time Delay
93(2)
6.4 Correlation Dimension
95(1)
6.5 False Nearest Neighbors
96(1)
6.6 Cao's Embedding Dimension
97(1)
6.7 Space-Time Separation
98(1)
6.8 Lyapunov Exponents
99(1)
6.9 Poincare Sections
99(1)
6.10 Recurrence Plot
100(3)
6.11 Predictive Chaotic Model: Global and Local Modeling
103(1)
6.12 Model Setup
104(5)
6.12.1 Univariate predictive chaotic model
104(3)
6.12.2 Multivariate predictive chaotic model
107(2)
6.12.3 Global model: Neural networks
109(1)
6.13 Model Results and Discussion
109(3)
6.14 K-fold Cross Validation
112(3)
6.15 Summary
115(2)
Chapter 7 Enhancements: Resolving Issues of High Dimensionality, Phase Errors, Incompleteness and False Neighbors
117(24)
7.1 Phase Space Dimensionality Reduction
117(5)
7.1.1 Introduction
117(1)
7.1.2 Problems of dimensionality
118(1)
7.1.3 Principal component analysis
119(1)
7.1.4 Reducing the phase space dimension
119(1)
7.1.5 Model results and discussion
120(2)
7.2 Phase Error Correction
122(6)
7.2.1 Introduction
122(1)
7.2.2 Data description
123(1)
7.2.3 Setting up the 1st standard predictive chaotic model
124(1)
7.2.3.1 Finding the proper time delay
125(1)
7.2.3.2 Estimating the appropriate embedding dimension
125(1)
7.2.3.3 Using the proper number of neighbors
125(1)
7.2.4 Setting up the 2nd model (predictive chaotic model and ANN model
126(1)
7.2.4.1 Predictive chaotic model
126(1)
7.2.4.2 ANN model
126(1)
7.2.5 Model results and discussion
127(1)
7.3 Building Predictive Chaotic Model from Incomplete Time Series
128(5)
7.3.1 Introduction
128(2)
7.3.2 Weighted sum of linear interpolations
130(1)
7.3.3 Bayesian PCA
130(1)
7.3.4 Cubic spline interpolation
130(1)
7.3.5 Model results and discussion
131(2)
7.4 Finding True Neighbors
133(5)
7.4.1 Euclidean distance method
133(1)
7.4.2 The new trajectory based method
134(2)
7.4.3 Model results and discussion
136(2)
7.5 Summary
138(3)
Chapter 8 Computational Intelligence in Identifying Optimal Predictive Chaotic Model
141(14)
8.1 Introduction
141(2)
8.2 Randomized Search Algorithms
143(3)
8.2.1 Grid search
143(1)
8.2.2 Genetic algorithm (GA)
143(2)
8.2.3 Adaptive cluster covering algorithm (ACCO)
145(1)
8.3 Case Study
146(1)
8.4 Model Setup
147(3)
8.4.1 Main experiment: predictive model for Hoek van Holland
147(1)
8.4.1.1 Grid search
147(1)
8.4.1.2 Randomized search
148(1)
8.4.2 Additional experiment: predictive model for the San Juan station
148(1)
8.4.2.1 Grid search
149(1)
8.4.2.2 Randomized search
150(1)
8.5 Model Results and Discussion
150(3)
8.6 Summary
153(2)
Chapter 9 Real-Time Data Assimilation Using Narx Neural Network
155(14)
9.1 Introduction
155(3)
9.2 NARX Neural Network
158(1)
9.2.1 Network Architecture
158(1)
9.2.2 Learning Algorithm
158(1)
9.3 NARX Data Assimilation
159(2)
9.4 Data Description
161(1)
9.5 Model Results and Discussion
161(5)
9.5.1 Estimating delay time and embedding dimension
161(3)
9.5.2 European operational storm surge models
164(1)
9.5.3 Chaotic storm surge models
164(1)
9.5.4 Data assimilation using NARX neural network
165(1)
9.6 Summary
166(3)
Chapter 10 Ensemble Model Prediction
169(14)
10.1 Introduction
169(1)
10.2 Principles of Ensemble Model Prediction
169(6)
10.2.1 Information-theoretic model selection
170(1)
10.2.2 Bayesian model averaging
171(3)
10.2.3 Ensembles with spatial information
174(1)
10.2.4 Machine learning: modular model
174(1)
10.3 Linear Prediction Combination
175(1)
10.4 Nonlinear Prediction Combination
176(2)
10.4.1 Dynamic averaging
176(1)
10.4.2 Dynamic neural networks
177(1)
10.5 Model Results and Discussion
178(3)
10.5.1 Global model
178(1)
10.5.2 Local model
178(2)
10.5.3 Dynamic averaging
180(1)
10.5.4 Dynamic neural networks
180(1)
10.6 Summary
181(2)
Chapter 11 Conclusions and Recommendations
183(8)
11.1 Main Conclusions
183(4)
11.2 Limitations and Recommendations
187(4)
References 191(10)
About the Author 201(2)
Scientific Publications 203(4)
Samenvatting 207
Michael Siek earned his B.Sc.degree in Mathematics from Airlangga University and B.Com. degree in Information Management from STIKOM Institute, both in 2000 and M.Sc. degree in Hydroinformatics from UNESCO-IHE, The Netherlands in 2003. He received his Ph.D. degree in Hydroinformatics from Delft University of Technology (TUDelft) and UNESCO-IHE in 2011 with the thesis entitled Predicting storm surges: chaos, computational intelligence, data assimilation, ensembles. Previously, he worked as a full-time lecturer at University of Surabaya and a visiting lecturer at Petra Christian University in the Faculty of Engineering and Faculty of Economics. His research has spanned a large number of disciplines, emphasizing data-driven and physically-based modelling, hydrological and coastal modelling, nonlinear dynamics and chaos theory, computational intelligence, optimization techniques, data mining, data assimilation, multi-model ensemble predictions with a wide range of real-life applications.
Pastovi nuoroda: https://www.kriso.lt/db/9780429088261_pe.html