Research On Identification And Control Of Nonlinear System Based On Support Vector Machine

Most of the processes are nonlinear in real life,the method of the traditional linear can't identify nonlinear system identification,and it is also difficult to implement the effective internal control.Aiming at the above problems,this paper designs the identification for nonlinear system and control strategy based on SVR,the main contents are as follows:Firstly,the identification and control of nonlinear systems are analyzed deeply,proposing propose the use of SVR for nonlinear system identification and control research.In order to obtain the relevant parameters of SVR automatically,a fuzzy differential evolution(FDE)algorithm is proposed to obtain the SVR parameters automatically.Secondly,the identification method of nonlinear system based on FDE-SVR is designed.In order to illustrate the superiority of the FDE algorithm,there are two experiments: first,using fuzzy differential evolution algorithm to obtain the relevant parameters automatically,and then use the parameters to conduct the nonlinear system identification;second,using the standard differential evolution algorithm automatically to obtain the relevant parameters,it conduct the nonlinear system identification next.The experimental results show that the nonlinear system identification method based on FDE-SVR can improve accuracy and speed of the identificationFinally,the internal model control of nonlinear system is realized based on FDE-SVR.First,using FDE-SVR,it can identify internal model M which exist in model control;and then it can identify inverse model C which exist in internal model control using FDE-SVR;finally,combining the internal model M and inverse model C which can obtain the system.By using the order step signal,sine wave,square wave and sawtooth signal,the tracking test based on the internal model control system of FDE-SVR can conduct.The results show that the control system can effective track aiming the input signal which are giving,and has strong robustness and Anti-jamming capability.