Intelligent Decoupling Control System of Complex Industrial Process and Application

Authors:	Chai Tianyou, Northeastern University, China Yue Heng, Northeastern University, China Zhai Lianfei, Northeastern University, China
Topic:	3.2 Cognition and Control ( AI, Fuzzy, Neuro, Evolut.Comp.)
Session:	Neural Control
Keywords:	Control applications; Multivariable control systems; Decoupling problems; Intelligent control; Switches; Neural networks; Distributed control systems

Abstract

A new decoupling control structure is presented for complex industrial processes, which consists of a feedback controller, a decoupling compensator and a feedforward compensator for unmodeled dynamics. Using this structure, an intelligent decoupling control system is established which realizes an adaptive decoupling control strategy for industrial processes with integrated complexities by using a neural network plus a switching mechanism. Such system can be easily implemented on distributed control systems (DCS) and has been successfully applied to ball mill pulverizing systems for 200MW power units. Simulation and industrial application show its robust performance of the proposed system and its potential prospect in the industry.