Research On Anti-Disturbance Control Method For Permanent Magnet Synchronous Motor Servo System Under Various Working Conditions

Permanent magnet synchronous motor(PMSM)servo system can meet the demands of high speed and high precision manufacturing,and is widely applied in industrial automation systems.However,in actual conditions,the PMSM servo system is always influenced by various uncertainties,such as time-varying internal parameters,unknown external load disturbances and nonlinear connections.Therefore,the PMSM servo system should own the excellent ability of both dynamic response and disturbance rejection.Supported by the projects of the National Natural Science Foundation of China,in this thesis,the mechanical structure and the vector control theory of PMSM is analyzed,and then the integrated disturbance model on the base of the internal uncertainties and external disturbances is established,furthermore,the evaluation index of the disturbance rejection is introduced.Next,three representative anti-disturbance control methods are selected to carry out research according to the complexity of the disturbance in working conditions,i.e.,the two degree-of-freedom PI parameters auto tuning control,the active disturbance rejection control,and the sliding mode control,respectively.The specific contents and innovations are listed as follows.In the case of simple disturbance,aiming at the anti-disturbance method based on the two degree-of-freedom PI parameters auto tuning control,an improved just-in-time learning algorithm is presented to identify the controlled model,using the search strategy named "rough search engine and refined search engine" to find the similar sample data,which has higher identification precision and modeling efficiency;the fractional order generalized predictive control is put forward to substitute for the integer order one as the performance evaluation index,improving the flexibility and effectiveness in tuning the two degree-of-freedom PI parameters;the control input of the two degree-of-freedom PI control is compared with that of the fractional order generalized predictive control to obtain the control parameters.In the case of medium disturbance,aiming at the anti-disturbance method based on the active disturbance rejection control,a tracking differentiator on the base of the proximate time optimal control is proposed,arranging the suitable transition for the command signal and extracting the appropriate differential signal,whose control parameters tuning is related with the damping ratio and the undamped natural frequency,and which is not sensitive to the discrete noise on the command signal;the variable damping nonlinear feedback controller based on the composite control is also presented,setting a smaller damping ratio in the tracking process in order to speed up the transient response,and setting a larger damping ratio during the steady-state process to reduce the overshoot;the reduced order extended state observer is designed to preserve the hardware resource and guarantee the real-time performance of the servo system without observing the state directly measured by sensors.In the case of complex disturbance,aiming at the anti-disturbance method based on the sliding mode control,a fast continuous terminal sliding mode control is presented,applying the exponential reaching law and fast terminal attractor to accelerate the convergence at the approaching stage and the sliding stage respectively,and introducing the extended state observer to feedforward compensate the integrated disturbance for reducing the switching gain and relieving the chattering phenomenon;the exponential adaptive integral sliding mode control is put forward which can be separated into the nominal part and the disturbance rejection part,in the nominal part using the proximate time optimal control to make the system state stabilization rapidly under the condition of saturation constraints,and in the disturbance rejection part adopting the exponential adaptive law to dynamically adjust the switching gain in the linear sliding mode control for decreasing chattering phenomenon and keeping enough robustness.The experimental platform of moment of inertia and load torque of PMSM servo system is set up,which includes the numerical control device,the servo driver,and the PMSM.The above three anti-disturbance methods are experimentally verified.The experimental results demonstrate the repeatability,the effectiveness and the robustness of the proposed disturbance rejection methods.