Available:*
Library | Item Barcode | Call Number | Material Type | Item Category 1 | Status |
---|---|---|---|---|---|
Searching... | 30000010133305 | TD886.5 K42 2007 | Open Access Book | Book | Searching... |
On Order
Summary
Summary
Artificial neural networks (ANNs), which are parallel computational models, comprising of interconnected adaptive processing units (neurons) have the capability to predict accurately the dispersive behavior of vehicular pollutants under complex environmental conditions. This book aims at describing step-by-step procedure for formulation and development of ANN based VP models considering meteorological and traffic parameters. The model predictions are compared with existing line source deterministic/statistical based models to establish the efficacy of the ANN technique in explaining frequent dispersion complexities in urban areas.
The book is very useful for hardcore professionals and researchers working in problems associated with urban air pollution management and control.
Table of Contents
1 Introduction | p. 1 |
1.1 Air Pollution Definition | p. 2 |
1.1.1 Composition of Atmosphere | p. 2 |
1.2 Air Pollution Problems | p. 3 |
1.3 Air Pollution Sources | p. 4 |
1.3.1 Point Source Emissions | p. 4 |
1.3.2 Area Source Emissions | p. 4 |
1.3.3 Line Source Emissions | p. 5 |
1.4 Urban Air Pollution Control Strategies | p. 5 |
1.5 Modelling Tools - Conventional and Soft Computational Approach Including ANN | p. 5 |
2 Vehicular Pollution | p. 7 |
2.1 General | p. 7 |
2.2 Sources of Vehicular Pollution | p. 8 |
2.3 Types of Vehicular Pollutants | p. 10 |
2.3.1 Carbon Monoxide | p. 10 |
2.3.2 Nitrogen Oxides | p. 10 |
2.3.3 Volatile Organic Compounds | p. 11 |
2.3.4 Sulphur Dioxide | p. 11 |
2.3.5 Particulate Matter | p. 12 |
2.3.6 Lead | p. 12 |
2.4 Health Effects of Vehicular Pollution | p. 13 |
2.5 Meteorological and Topographical Factors Affecting Vehicular Pollution Dispersion in Urban Air Sheds | p. 15 |
2.6 Ambient Air Quality Monitoring | p. 18 |
2.7 Local Air Quality Management | p. 19 |
2.8 Options for Control of Vehicular Pollution | p. 22 |
2.9 Ambient Air Quality Standards | p. 23 |
2.10 Overview of Vehicular Pollution Modelling | p. 23 |
3 Artificial Neutral Networks | p. 25 |
3.1 General | p. 25 |
3.2 What Artificial Neural Networks are? | p. 25 |
3.3 Basic Concepts of Neural Network | p. 26 |
3.3.1 Human Biological Neuron | p. 26 |
3.3.2 Simple Neuron Model | p. 28 |
3.4 History of Artificial Neural Network | p. 29 |
3.5 Artificial Neural Network Architecture | p. 30 |
3.6 Types of Neural Networks | p. 31 |
3.6.1 Feed-Forward Networks | p. 32 |
3.6.2 Recurrent Neural Networks | p. 32 |
3.7 Transfer Functions and Learning Algorithms | p. 34 |
3.7.1 Transfer Functions | p. 34 |
3.7.2 Learning Methods | p. 34 |
3.8 Back-Propagation Learning Algorithm | p. 35 |
3.9 Summary | p. 39 |
4 Vehicular Pollution Modelling-Conventional Approach | p. 41 |
4.1 General | p. 41 |
4.2 Theoretical Approaches of Vehicular Pollution Modelling | p. 42 |
4.3 Vehicular Pollution Deterministic Models | p. 47 |
4.4 Vehicular Pollution Numerical Models | p. 55 |
4.5 Vehicular Pollution Stochastic Models | p. 58 |
4.6 ANN based Vehicular Pollution Models | p. 61 |
4.7 Limitations of Vehicular Pollution Models | p. 63 |
4.8 Summary | p. 66 |
5 Vehicular Pollution Modelling - ANN Approach | p. 67 |
5.1 General | p. 67 |
5.2 ANN Approach to Vehicular Pollution Modelling | p. 68 |
5.3 Algorithm for ANN based Vehicular Pollution Model | p. 69 |
5.3.1 Selection of the Optimal ANN based Vehicular Pollution Model Architecture | p. 70 |
5.3.2 Selection of the Best Activation Functions | p. 71 |
5.3.3 Selection of the Optimum Learning Parameters | p. 71 |
5.3.4 Initialization of the Network Weights and Bias | p. 72 |
5.3.5 Training Procedure | p. 73 |
5.4 Statistics for Testing ANN based Vehicular Pollution Models | p. 77 |
5.5 Development of ANN based Vehicular Pollution Models | p. 78 |
5.6 Case Study | p. 79 |
5.6.1 Pollutant Data | p. 81 |
5.6.2 Traffic Data | p. 84 |
5.6.3 Meteorological Data | p. 85 |
5.6.4 Models Development | p. 86 |
5.7 Summary | p. 119 |
6 Application of ANN based Vehicular Pollution Models | p. 121 |
6.1 General | p. 121 |
6.2 Model Performance Indicators | p. 122 |
6.2.1 Root Mean Square Error | p. 122 |
6.2.2 Coefficient of Determination | p. 123 |
6.2.3 Mean Bias Error | p. 124 |
6.2.4 Standard Deviations | p. 124 |
6.2.5 Slope and Intercept of the Least Square Regression Equation | p. 125 |
6.2.6 Degree of Agreement | p. 125 |
6.3 Application of ANN Based Vehicular Pollution Models at Urban Intersection and Straight Road Corridor | p. 125 |
6.3.1 1-hr Average CO Models | p. 125 |
6.3.2 8-hr Average CO Models | p. 133 |
6.3.3 24-hr Average NO[subscript 2] Models | p. 140 |
6.4 Performance Evaluation and Comparison of ANN based Vehicular Pollution Models with Conventional Models | p. 147 |
6.4.1 Performance of ANN based CO Models for the Critical Period Test Data | p. 147 |
6.4.2 Performance of Univariate Stochastic Models for the Critical Period Test Data | p. 149 |
6.4.3 Performance of Deterministic Model for the Critical Period Test Data | p. 151 |
6.5 Summary | p. 155 |
7 Epilogue | p. 157 |
Appendix A p. 163 | |
Appendix B p. 175 | |
Appendix C p. 185 | |
Appendix D p. 211 | |
References | p. 227 |