Research On Parameter Identification And Intelligent Control Method Of Permanent Magnet Synchronous Moto

With the implementation of the "Made in China 2025" strategy,intelligent manufacturing has become an inevitable trend and direction for the development of China’s manufacturing industry.Particularly in the aerospace field,Permanent Magnet Synchronous Motors(PMSM)have gained significant attention as a crucial component of the propulsion system.In order to improve the overall performance of PMSM system,the research work is carried out from two aspects,one is parameter identification method based on adaptive forgetting factor and auxiliary variables,the other is maximum entropy soft actor-critic(SAC)intelligent control method.The main work is as follows:(1)To address the issue of parameter variation in PMSM under low and high temperature alternating conditions,this thesis proposes a recursive least squares method based on adaptive forgetting factor and instrumental variables by analyzing the properties of forgetting factors and instrumental variables.The method utilizes a dynamic adaptive forgetting factor adjustment function to balance the influence of new and old data on the identification results of motor parameters by predicting the error between the predicted output value of the motor model and the true value.At the same time,instrumental variables are added to solve the problem of biased identification results caused by colored noise in system data.Simulation results show that compared with the recursive least squares method with a fixed forgetting factor,the proposed algorithm has better convergence ability and stability under the adjustment of adaptive forgetting factor when the resistance,inductance,and magnetic flux of PMSM change abruptly.Meanwhile,the added instrumental variables can ensure unbiased identification results.(2)To address the problems of time-consuming and laborious manual adjustment of parameters and the instability of PMSM systems caused by PID control methods under sudden load torque changes,this thesis proposes a selflearning maximum entropy SAC intelligent control algorithm that uses data generated by the interaction between PMSM intelligent agents and the environment as the driving force.The intelligent control algorithm replaces the q-axis speed loop and current loop in traditional PID control with direct output of q-axis voltage,realizing the intelligent control of PMSM.Simulation results show that under variable load and variable speed conditions,compared with traditional PID control and DDPG intelligent control methods,the maximum entropy SAC intelligent control method effectively improves the system’s antiinterference ability,tracking performance and stability.