Controller based of fuzzy wavelet neural network for control of technological processes

Abstract

In industry some technological processes are characterized by nonlinearity of functioning, uncertainty of environment. One of main problem for effective control of these processes is the creating of the proper knowledge base for the controller. In this paper the integration of fuzzy set theory and wavelet neural network (WNN) is considered to solve this problem. The structure and operation algorithms of fuzzy WNN based controller are presented. Using gradient method the learning of fuzzy WNN is performed to find optimal values of the parameters of controller. The simulation of fuzzy WNN based control system for control of dynamic plant is carried out. Result of simulation of control system based on fuzzy WNN is compared with the simulation result of control systems based on feedforward neural network and PID controller. Simulation results demonstrate that training of fuzzy WNN based control system is faster and it has better control performance than others.