Research On Parameter Identification Of Permanent Magnet Synchronous Motor Based On Neural Network

The speed and power frequency of three phase AC permanent magnet synchronous motor are kept strictly synchronized, with the advantages of high efficiency, fast dynamic response, high reliability and so on. Meanwhile, with the continuous decline of the price of permanent magnetic material and the continuous improvement of the performance, the permanent magnet synchronous motor has achieved a wide range of applications in the field of high performance motion control.The rapid development of modern high performance numerical control machine tools and electric vehicles require that the drive system has higher precision and better control performance, thus the requirements of high precision and high control performance of permanent magnet synchronous motor are proposed. The design method of permanent magnet synchronous motor controller is generally required to control system parameters precisely so that the control law can be adjusted in real time. However, in the actual operation of the system, the parameters of the system shall always change. To ensure the excellent control performance, the controller must be adjusted accordingly. Therefore, various identification algorithms are studied to identify the unknown parameters of the motor system.Direct current attenuation method, least square method, model reference adaptive method and other parameters identification methods have been successfully applied in the field of parameter identification of permanent magnet synchronous motor, but there are some limitations in these control methods.According to the current situation of time variation of the electrical parameters of permanent magnet synchronous motor, the intelligent control algorithm based on neural network is used to identify the electrical parameters of permanent magnet synchronous motor. The specific research contents and main innovations include the following aspects:1. The methods of parameter identification are summarized in this paper, including the least square method, the model reference adaptive method, the extended kalman filtering method and various intelligent control algorithms. At the same time, the relationship between the electrical parameters of the motor and the controller parameters of the double closed loop speed regulation system is presented, and the importance of motor parameter identification is presented. The mathematical model of permanent magnet synchronous motor in different coordinates and the control strategy of permanent magnet synchronous motor are described in this paper. The space vector pulse width modulation technology based on vector control is described.2. The research and analysis of the algorithm of recursive least squares parameter identification with a forgetting factor is given. The differences of motor parameter identification results are given under different forgetting factors. Ample theoretical and experimental researches are carried out to analyze parameter identification technique of permanent magnet synchronous motor based on neural network. The reference model and adjustable model are built depend on the motor mathematical model. The stator resistance of the motor, the direct axis inductance and the flux of permanent magnet are identified by the continuous adjustment of the weights, and the feasibility of the identification method is verified by MATLAB.3. The influence of the nonlinear factor of inverter on the identification results is analyzed and a reasonable compensation method is given. Parameter identification of permanent magnet synchronous motor with the nonlinear factor of inverter is simulated, and it is necessary to compensate the nonlinear factors in the practical application according to the identification results, so as to improve the accuracy of parameter identification.