Double Closed Loop Optimization Of Permanent Magnet Synchronous Motor Based On MRAS Observer

The interior permanent magnet synchronous motor has been widely used in CNC machine tools,aerospace,new energy electric vehicles and other servo speed control systems,dut to its characteristics of high power density,large starting torque,easy to weak magnetic speed expansion,easy to control,simple structure,etc.At present,vector control method often used for permanent magnet synchronous motor can control the accurate torque and rotational speed;However the high cost of mechanical sensors in the traditional motor vector control system,cumbersome installation,the environmental effect will limit its scope of use,at the same time reduce the reliability of the system.So the sensorless speed control strategy research is now hot spot all over the world.In vector speed sensorless control system,both speed loop and current loop with the traditional fixed PI parameters are difficult to meet complex working conditions and the torque pulsation is large.Aiming at the above problems,this paper mainly carries out double closed-loop optimization design in speed loop and current loop.That is to improve the speed observer accuracy,speed regulation performance,torque ripple suppression in three aspects.Firstly,the rotational speed observer in the outer speed loop is the nonlinear timevarying feedback link of the Model Reference Adaptive System(MRAS),the traditional method is to use Popov’s superstability theorem to derive an adaptive law in the form of proportional integral to determine the rotational speed.The observed speed convergence is slow,the accuracy is poor,and the anti-interference ability is weak.Therefore,the sliding mode variable structure is used to replace the proportional integral controller in the adaptive law to enhance the robustness of the system,and the stability of the improved adaptive law is proved by Lyapunov Stability Theorem.Since a new sliding mode variable structure model reference adaptive observer was obtained to combine the adaptive law with the sliding mode variable structure,which brought both robustness and jitter,and the jitter problem was optimized.The performance of the observed speed is determined by the speed regulator of the speed loop.The speed regulator also widely adopts the ordinary proportional integral controller,which has the contradiction between the starting performance and antiinterference ability.In order to solve this contradiction,a speed sliding mode controller based on load disturbance observation is designed to replace the traditional proportional integral controller in the speed loop.The simulation results show that the motor’s starting performance and anti-interference ability have been significantly improved.Secondly,in order to suppress torque pulsation and reduce current ripple in steady-state,a model predictive control algorithm with limited current set was designed to replace the current loop,which can not only reduce PI parameter setting and complex SVPWM modulation,but also optimize the internal current loop.To some extent,the complexity of the model is simplified and the torque fluctuation is suppressed.The simulation results show that the new dual-sliding mode MRAS model predictive control system optimized by speed loop and current loop has improved the motor speed observation accuracy and speed regulation performance,and the torque ripple has been significantly suppressed compared with the traditional MRAS vector control system.Finally,a servo hardware test platform has been built with design of the algorithm and the upper computer software of the dual sliding mode MRAS model predictive control.And the experimental verification and comparison are carried out under different interference conditions.The experimental results show that the new dual sliding mode improved by the double closed loop compared with the traditional MRAS vector control system,the MRAS model predictive control system has better stability and stronger antiinterference ability.