Kayacan E., Khanesar M.A. Fuzzy Neural Networks for Real Time Control Applications. Concepts, Modeling and Algorithms for Fast Learning

Butterworth-Heinemann, 2016. — 254 p.This book presents the basics of FNNs, in particular T2FNNs, for the identification and learning control of real-time systems. In addition to conventional parameter tuning methods, e.g., GD, SMC theory-based learning algorithms, which are simple and have closed forms, their stability analysis are also introduced. This book has been prepared in a way that can be easily understood by those who are both experienced and inexperienced in this field. Readers can benefit from the computer source codes for both identification and control purposes that are given at the end of the book.
There are number of books in the area of FLSs and FNNs. However, this book is more specific in several aspects. First of all, whereas so many books focus on the theory of type-1 and type-2 FLCs, we give more details on the parameter update algorithms of FNNs and their stability analysis. Second, the emphasis here is on the SMC theory-based learning algorithms for the training of FNNs, because we think these algorithms are the simplest and most efficient methods when compared to other algorithms, e.g., the GD algorithm. Last but not least, this book is prepared from the view of the identification and control of real-time systems, which makes it more practical.
The fuzzy logic principles were used to control a steam engine by Ebraham Mamdani of University of London in 1974. It was the first milestone for the fuzzy logic theory. The first industrial application was a cement kiln built in Denmark in 1975. In the 1980s, Fuji Electric applied fuzzy logic theory to the control a water purification process. As a challenging engineering project, in 1987, the Sendai Railway system that had automatic train operation FLCs since from 1987, not many books are available in the market as a reference for real-time systems. This book aims at filling this gap.Mathematical Preliminaries
Fundamentals of Type-1 Fuzzy logic Theory
Fundamentals of Type-2 Fuzzy logic Theory
Type-2 Fuzzy Neural Networks
Gradient Descent Methods fur Type-2 Fuzzy Neural Networks
Extended Kalman Filter Algorithm fur the Tuning of Type-2 Fuzzy Neural Networks
Sliding Mode Control Theory-Based Parameter Adaptation Rules fur Fuzzy Neural Networks
Hybrid Training Method fur Type-2 Fuzzy Neural Networks Using Particle Swarm Optimization
Noise Reduction Property of lYpe-2 Fuzzy Neural Networks
Case Studies: Identification Examples
Case Studies: Control Examples