Research On Self-tuning Strategy Of Auto-disturbance Rejection Control Parameters For Pmsm

Permanent magnet synchronous machine(PMSM)is widely used in high-performance servo systems because of their simple structure,high power density,and fast response speed.However,in actual industrial applications,the performance of the motor drive system is often affected by internal and external disturbances.And there is a decline.In order to quickly and effectively suppress various disturbances and improve the performance of the permanent magnet AC servo system,this paper adopts the Active Disturbance Rejection Controller(ADRC)to improve the design of the servo system,and the selection of parameters is important for the control performance of the Active Disturbance Rejection Controller.Influence,starting from the operating conditions of the system,a set of parameter self-tuning and optimization methods are proposed.First,in order to obtain the theoretical basis for the design of the PMSM servo system,the mathematical models of the surface mount permanent magnetic synchronous machine(SPMSM)in different coordinate systems are derived in detail,and the vector control strategy with id = 0and the realization process of SVPWM are explained.And analyze the difference between linear and non-linear ADRC based on the principle and mathematical model of ADRC.Secondly,in order to achieve decoupling control and enhance the robustness of the control system,a double closed-loop auto-disturbance-rejection vector control method for speed and current is proposed,and the control performance and controller design complexity are considered comprehensively,and the PMSM vector control system is specifically designed.The current loop and the rotational speed loop ADRC,give the method of tuning some parameters of the controller.Then,in view of the problem that the nonlinear ESO parameter tuning is complex and has a great impact on the control performance,the control performance evaluation index of the servo system is studied,and the off-line auto-tuning method of the auto disturbance rejection parameter of the permanent magnet synchronous motor based on the genetic algorithm(GA)and NSGA-Ⅱ algorithm is proposed.,Through the iterative evolution of the population under multiple objective functions to find the optimal ADRC parameters,the comprehensive evaluation of the performance of the PMSM servo system is realized.The GA single-objective and NSGA-Ⅱ multi-objective offline parameter auto-tuning are compared and verified with the traditional tuning method,which proves the effectiveness and convergence of the method designed in this paper.Finally,aiming at the problem that the control performance of the servo system with fixed parameters is degraded when the working status changes,an on-line auto-tuning method of auto disturbance rejection parameters of permanent magnet synchronous motor based on BP neural network is proposed.This method learns the system performance in real time according to the arbitrary nonlinear expression ability of BP neural network,adjusts its own weight continuously,and replaces the fixed parameters with the adaptive and dynamically updated auto-disturbance rejection controller parameters obtained from the network output.It focuses on the determination of the network structure,the selection of the initial value of the weight and the selection of the activation function.An experimental platform based on d SPACE is built to verify the method,and the results show that the online tuning of parameters using BP neural network makes the system have better anti-disturbance performance.