Based On Dsp Belt Conveyor Motor Drive Power Balance Fuzzy Controller

For the multi-motor driving belt system, the control of each motor's power balance is a multi-input multi-output (MIMO) and nonlinear complex system. The conventional linear control method based on the precise mathematic model can hardly to get required control effect While fuzzy control do not need to know accurate mathematic model of object, but is based on human thinking and producing experience; and because it applies language rule to describe controlling process and bases rule to modify the controlling arithmetic and parameter, at the same time these two mathematic has self-adaptability, they will realize production process control.MIMO power balance fuzzy control of multi-motor driving belt system is studied in this dissertation. The important item of designing multivariable fuzzy controller is to obtain the rules of the multivariable fuzzy control. Therefore, the rules are generated with self-adapting, self-organization and self-learning. Neural network has the strong capacity of self-adapting and self-learning and it can be well combined with fuzzy control.Multivariable fuzzy neutral network control is applied in multi-motor driving belt power-balance control system in this research. Based on the analysis of fuzzy and neural network control, a self-organized fuzzy neural network controller is proposed. This controller has the capacity of constructing and parameter learning. It needs no pre-determined rules and it can learn the rules online from sample data through the process of control. Moreover, it can adjust the structure of net, produce the rules, and regulate the parameters.Fuzzy toolbox of MATLAB is used in this paper to emulate the system. The analysis of simulation result shows that the present fuzzy neural network control has the features of rapid response speed, small overshoot, slight steady-state error and robust stability.Finally in view of the multi-motor driving belt power-balance control system, DSP(digital signal processor) fuzzy neural controller is developed. The design of software and hardware of DSP have been finished. DSP technology and fuzzy neural network structure presented in the paper pave the way for the engineering application of fuzzy neural network controller.