DNA Computing Based Genetic Algorithm: Applications in Industrial Process Modeling and Control

Description

This book focuses on the implementation, evaluation and application of DNA/RNA-based genetic algorithms in connection with neural network modeling, fuzzy control, the Q-learning algorithm and CNN deep learning classifier. It presents several DNA/RNA-based genetic algorithms and their modifications, which are tested using benchmarks, as well as detailed information on the implementation steps and program code. In addition to single-objective optimization, here genetic algorithms are also used to solve multi-objective optimization for neural network modeling, fuzzy control, model predictive control and PID control. In closing, new topics such as Q-learning and CNN are introduced. The book offers a valuable reference guide for researchers and designers in system modeling and control, and for senior undergraduate and graduate students at colleges and universities. Contents 1 Introduction 1.1 Standard Genetic Algorithm 1.2 State of Art for GA 1.2.1 Theoretical Research of GA 1.2.2 Encoding Problem of GA 1.2.3 Constraint Handling in GA 1.2.4 Multi-objective Genetic Algorithm 1.2.5 Applications of GA 1.3 DNA Computing Based GA 1.3.1 DNA Molecular Structure of DNA Computing 1.3.2 Biological Operators of DNA Computing 1.3.3 DNA Computing Based Genetic Algorithm 1.4 The Main Content of This Book References 2 DNA Computing Based RNA Genetic Algorithm 2.1 Introduction 2.2 RNA-GA Based on DNA Computing 2.2.1 Digital Encoding of RNA Sequence 2.2.2 Operations of RNA Sequence 2.2.3 Encoding and Operators in RNA-GA 2.2.4 The Procedure of RNA-GA 2.3 Global Convergence Analysis of RNA-GA 2.4 Performance of the RNA-GA 2.4.1 Test Functions 2.4.2 Adaptability of the Parameters 2.4.3 Comparisons Between RNA-GA and SGA 2.5 Summary Appendix References 3 DNA Double-Helix and SQP Hybrid Genetic Algorithm 3.1 Introduction 3.2 Problem Description and Constraint Handling 3.3 DNA Double-Helix Hybrid Genetic Algorithm (DNA-DHGA) 3.3.1 DNA Double-Helix Encoding 3.3.2 DNA Computing Based Operators 3.3.3 Hybrid Genetic Algorithm with SQP 3.3.4 Convergence Rate Analysis of DNA-DHGA 3.4 Numeric Simulation 3.4.1 Test Functions 3.4.2 Simulation Analysis 3.5 Summary Appendix References 4 DNA Computing Based Multi-objective Genetic Algorithm 4.1 Introduction 4.2 Multi-objective Optimization Problems 4.3 DNA Computing Based MOGA (DNA-MOGA) 4.3.1 RNA Encoding 4.3.2 Pareto Sorting and Density Information 4.3.3 Elitist Archiving and Maintaining Scheme 4.3.4 DNA Computing Based Crossover and Mutation Operators 4.3.5 The Procedure of DNA-MOGA 4.3.6 Convergence Analysis of DNA-MOGA 4.4 Simulations on Test Functions by DNA-MOGA 4.4.1 Test Functions and Performance Metrics 4.4.2 Calculation Results 4.5 Summary Appendix References 5 Parameter Identification and Optimization of Chemical Processes 5.1 Introduction 5.2 Problem Description of System Identification 5.2.1 Lumping Models for a Heavy Oil Thermal Cracking Process 5.2.2 Parameter Estimation of FCC Unit Main Fractionator 5.3 Gasoline Blending Recipe Optimization 5.3.1 Formulation of Gasoline Blending Scheduling 5.3.2 Optimization Results for Gasoline Blending Scheduling 5.4 Summary Appendix References 6 GA-Based RBF Neural Network for Nonlinear SISO System 6.1 Introduction 6.2 The Coke Unit 6.3 RBF Neural Network 6.4 RNA-GA Based RBFNN for Temperature Modeling 6.4.1 Encoding and Decoding 6.4.2 Fitness Function 6.4.3 Operators of RBFNN Optimization 6.4.4 Procedure of the Algorithm 6.4.5 Temperature Modeling in a Coke Furnace 6.5 Improved MOEA Based RBF Neural Network for Chamber Pressure 6.5.1 Encoding of IMOEA 6.5.2 Optimization Objectives of RBFNN Model 6.5.3 Operators of IMOEA for RBFNN 6.5.4 The Procedure of IMOEA 6.5.5 The Chamber Pressure Modeling in a Coke Furnace 6.6 PCA and INSGA-II Based RBFNN Disturbance Modeling of Chamber Pressure 6.6.1 RV Criterion in PCA Variable Selection 6.6.2 Encoding of RBFNN 6.6.3 Operators of INSGA-II 6.6.4 The Procedure of Improved NSGA-II 6.6.5 Main Disturbance Modeling of Chamber Pressure 6.7 Summary References 7 GA Based Fuzzy Neural Network Modeling for Nonlinear SISO System 7.1 Introduction 7.2 T-S Fuzzy Model 7.2.1 T-S Fuzzy ARX Model 7.2.2 T-S Fuzzy Plus Tah Function Model 7.3 Improved GA based T-S Fuzzy ARX Model Optimization 7.3.1 Hybrid Encoding Method 7.3.2 Objectives of T-S Fuzzy Modeling 7.3.3 Operators of IGA for T-S Fuzzy Model 7.3.4 Optimization Procedure 7.3.5 Computing Complexity Analysis 7.3.6 Oxygen Content Modeling by Fuzzy ARX Model 7.4 IGA Based Fuzzy ARX Plus Tanh Function Model 7.4.1 Encoding of IGA for Fuzzy Neural Network 7.4.2 Operators of IGA for New Fuzzy Model 7.4.3 Liquid Level Modeling by Nonlinear Fuzzy Neural Network 7.5 Summary References 8 PCA and GA Based ARX Plus RBF Modeling for Nonlinear DPS 8.1 Introduction 8.2 DPS Modeling Issue 8.2.1 Time/Space Separation via PCA 8.2.2 Decoupled ARX Model Identification 8.2.3 RBF Neural Network Modeling 8.2.4 Structure and Parameter Optimization by GA 8.2.5 Encoding Method 8.3 Simulation Results 8.3.1 Catalytic Rod 8.3.2 Heat Conduction Equation 8.4 Summary References 9 GA-Based Controller Optimization Design 9.1 Introduction 9.2 Non-minimal State-Space Predictive Function PID Controller 9.2.1 Process Model Formulation 9.2.2 PID Controller Design 9.2.3 GA-Based Weighting Matrix Tuning 9.2.4 The Chamber Pressure Control by PFC-PID 9.3 RNA-GA-Based Fuzzy Neuron Hybrid Controller 9.3.1 Neuron Controller 9.3.2 Simple Fuzzy PI Control 9.3.3 Fuzzy Neuron Hybrid Control (FNHC) 9.3.4 Parameters Optimization of RNA-GA 9.3.5 Continuous Steel Casting Description 9.3.6 FNHC Controller Performance Analysis 9.4 Stabilization Subspaces Based MOGA for PID Controller Optimization 9.4.1 Generalized Hermite-Biehler Theorem 9.4.2 Hermite-Biehler Theorem Based PID Controller Stabilizing 9.4.3 Optimizing PID Controller Parameters Based on Stabilization Subspaces 9.4.4 Simulation for Optimization of PID Controllers 9.5 Summary References 10 Further Idea on Optimal Q-Learning Fuzzy Energy Controller for FC/SC HEV 10.1 Introduction 10.2 FC/SC HEV System Description 10.3 Q-Learning Based Fuzzy Energy Management Controller 10.3.1 Fuzzy Energy Management Controller 10.3.2 Q-Learning in HEV Energy Control 10.3.3 GA Optimal Q-Learning Algorithm 10.3.4 Initial Value Optimization of Q-Table 10.3.5 Procedure of Improved Q-Learning Fuzzy EMS 10.3.6 Real-Time Energy Management 10.4 Summary References