Neural Network Identification Method Applied to the Nonlinear System
Bo Liu, Huiguang Li, WU Ti-hua · 2009
A kind of neural network solution method has been proposed in the paper aiming at a class of non-linear process control system with the characteristic of time delay. In this scheme, a new-type associative memory neural network is used to model the controlled system, and the fuzzy neural network with inverse identification structure is adopted to control the nonlinear process system. This fuzzy neural network control method adopts the structure of three layers combine neural network identifier with inverse structure. Computer simulation and lab application show that it is effective to adopt this scheme to control on-linear process system with time delay.