The Research Of Algorithm By Parameters Self-Tuning Based On Fractional Order Controller

With the rapid development of computer technology in recent years, the calculation ofthe fractional calculus has become possible; the application of fractional calculus has openedup a vast space in various engineering applications and has been greatly enhanced. Meanwhile,the control of fractional controller and intelligent control algorithm combining is becomingmore and more attention.Nowadays, intelligent control methods include fuzzy control methods, geneticalgorithms, the controller of neural networks and expert controller. What we discussed is acombination of neural network intelligent algorithm with fractional controller to control themotor speed, the first step is to discredited the fractional operator, then adjust theparameters of the fractional order controller by using BP neural network algorithm in order toachieve the optimal control purposes.This paper is about the problem of fractional order controller based on BP neuralnetwork parameters tuning, specifically the following several parts:(1)To introduce the basic theory of fractional calculus and its correlation functionsystematically and in detail, and introduce a variety relationship of the functional betweenfractional calculi.(2)Descript the integer order PID controller algorithm specifically, such as theincremental algorithm and location algorithm; then introduce the control algorithm of thefractional order controller.(3)Overview the basic principles and characteristics of the neural network and BPalgorithm detailed instructions; with BP algorithm, parameter controls to adjust the fractionalcontroller.(4)Compare the results of effect between the two controller of integer order PIDcontroller and the fractional order controller based on the BP neural network control, throughthe constant speed control experiments and constant voltage control experiment, and byanalysis the different motor control effect of the fractional order controller of the BP neuralnetwork in different motor, then analysis the experimental results and draw conclusions.The innovation of this paper is on the fractional order controller using the BP neuralnetwork intelligent algorithm for processing, and specific implementation of the threeparameters to be controlled, in order to achieve optimal control. Experiments show thatcomparing to with the integer order PID controller, not only in the anti-interference ability butalso goals that has the ability to follow the superior characteristics, and charged with thetarget to change its control of the effect. the Fractional Order Controller Based on BP neural networkable to adapt to external changes in circumstances, and make the appropriateadjustments.