Reseach Of Neural-network Decoupling Control System Based On DeviceNet Fieldbus

In the process of modern complex industrial production, the coupling problem between controlled object is a common phenomena, and so to research the decoupling control strategy of MIMO system and in use for process control, has important operation significance for improving produce efficiency ensuring running safely.This paper focus on decoupling control based on Neural Network. Direct against the strong coupling characteristic of double-container object in Rockwell Laboratory, introduct Fuzzy self-adaptive PID control and Neural Network into the design of the control system. Present a decoupling arithmetic, and modeling the double-container system by using Neural Network,The arithmetic composed of Fuzzy self-adaptive PID controller and dispersive decoupling network which is applied in the design of double-container control system to decouple this system. The simulation results and real-time control show that the designed system dispel the coupling between the systematic variables basically, it improved greatly the dynamic and static responsible specifications of the system. Through simulation and real-time control, Compared with the traditional PID controller, this intelligent control algorithm can achieve better control effect, and the control effect with Neural Network Decoupler is better than without it.