Research And Implementation On Parameters Auto-tuning Technology Of Servo Driver Device

PID is the most widely used control algorithm in servo drive device,however,the optimal parameters of PID controller depend on various factors,so it is not a easy job to tune the PID controller.Traditionally,the parameters are tuned manually.But the procedure is really complicated and time-consuming.Also,the parameters need to be rectified when the work condition change.Given this,research on the parameters auto-tuning technology of servo driver device was conducted with support from national science and technology important project,High Performance CNC System and Foundational Manufacturing Equipment and a parameters auto-tuning scheme was implemented on configurable servo driver device developed by our lab.The model of PMSM was established based on vector transformation theory.SVPWM modulation technology and PID control theory were described in detail.After that,the structure and control method of PMSM field oriented control were analyzed.Then the models of current loop and speed loop were obtained respectively.Two current sampling methods and a back electromotive force compensation method were put forward.Referring to the principle of automation,the PI controller parameters tuning formulas were calculated.Because the parameters of PMSM were unknown in the formulas,the identification algorithms of them were also studied.The simulation experiments were conducted using MATLAB/Simulink.In addition,considering the limitation of the control models,the parameters were further optimized by genetic algorithm.The design and implementation procedure of the genetic algorithm were described in detail.The algorithm was also verified in MATLAB/Simulink.Finally,the components that encapsulated the back electromotive force compensation algorithm,inertial identification algorithm and genetic algorithm were designed on the control software platform in order to tune the parameters automatically.The experimental results show that the proposed scheme is correct and practical.The method that the parameters are tuned based on the model and then optimized by genetic algorithm to obtain the optimal performance combines the normal model-based algorithm with the rule-based intelligent search algorithm,which provides a new idea for research on parameters auto-tuning technology of servo driver device.