Research On Adaptive Control For PMSM Based On Inertia Identification

Permanent magnet synchronous motor(PMSM)is widely used as actuator in motion control applications due to its high performance such as high power density,compact structure,high torque to inertia ratio and high efficiency.Many traditional linear control methods cannot achieve a sufficiently high performance for the PMSM system because of nonlinearities exist in the model and parameter variations caused by external factors.For example,the PI control method,which is simple inplementation and widely used,has the problem of long settling time,big overshoot and maladjustment when load change.Firstly,on the basis of the principle and structure of permanent magnet synchronous motor,combined with the coordinate transformation,the PMSM mathematical models under different coordinate system are derived and analyzed.Then the different current control strategies of vector control method are compared and i_d=0 plan is adopted in consideration of experimental motor's characteristics.Meanwhile,the features of common pulse width modulation methods are analyzed in the paper and SVPWM method with a high utilization of DC supply voltage is selected.Secondly,a composited controller combing the extended state observer(ESO)with Current Predictive Control(CPC)together(CPC-ESO)for the PMSM speed-regulation system is developed.Compared with PI control,CPC can obtain the voltage reference directly and as a consequence,a higher dynamic response performance.ESO,which inherits from classical PID and does not require accurate model parameters,has the characteristic of estimating both state variables and the lumged disturbances.With the disturbances observed by ESO,a feed forward compensation can be employed to develop auto disturbance rejection control.Thirdly,an adaptive control method with inertia identification and parameter self-tuning is proposed to handle the performance degradation caused by variations of load inertia.The equivalent moment of inertia was identified out to tune the gain of ESO feedforward coefficient.Finally,a MATLAB simulation model and a DSP experimental platform were built to verify the control performance of the proposed methods.In contrast to PI control in both current and speed loop,the adaptive speed control method composited in this paper possesses the advantages of good disturbances rejection,high dynamic performance and adaptation to inertia variations.