Optimal Design Of PMSM Speed Control System Based On Data Drive

In the double closed-loop control system of permanent magnet synchronous motor,the traditional linear control algorithm or nonlinear control algorithm is used to design the speed loop.The effectiveness of speed tracking is governed by the accuracy of the controlled object model and the adaptability of the regulator.Due to the rapid update of big data,cloud computing,machine learning and other technologies,more and more intelligent algorithms are applied to the field of motor control.Based on this,this paper studies a data-driven intelligent control algorithm for permanent magnet synchronous motor.And the algorithm was used to devise a speed regulator and applied to a motor speed control system.The algorithm is not constrained by the model of the controlled object and continuously manages the regulator through data-driven training,using the regulator’s self-learning capability to avoid the interference of non-linear factors such as parameter variability and dynamic disturbances.The excellence of the algorithm is evaluated using MATLAB simulation and a semi-physical simulation platform.The specific contents of this paper are as follows:(1)This paper introduced the characteristics,performance and application fields of different types of motors and compared the strategies of constant voltage frequency ratio,vector control and direct torque control.this paper also provides a brief review of regulators using PI,sliding film control and self-anti-disturbance design;and summaries the state of research on adaptive dynamic planning in intelligent control algorithms.(2)According to the position of permanent magnets,This paper compares the characteristics and performance of various types of permanent magnet motors,establishes the mathematical model in the natural coordinate and introduces coordinate transformation and finishes the model derivation.In addition,this paper analyzes several commonly used control methods in vector control,selects_di=0 as the control method,builds the motor speed control system model with PI regulator,and discusses the principle of voltage space vector pulse width modulation algorithm in the model,which provides the basis for the simulation and experiment.(3)According to the analysis and comparison of adaptive dynamic programming algorithm,the critical network and the execution network are designed by ADHDP algorithm to form an intelligent speed regulator.The control strategy of the system can be improved by data-driven,so as to improve the influence of motor model error and external nonlinear interference on the control performance of motor speed regulation system.In order to verify the effectiveness of the proposed method,the simulation results in MATLAB are given comparison of the proposed ADHDP controller and PI controller.(4)In order to verify the timeliness of the algorithm in the real environment,this paper built the hardware in the loop simulation experiment to analyze the control performance of the algorithm.Based on the MATLAB simulation model of ADHDP regulator,the bottom module of the model is built,and the environment of code generation is configured.The hardware in the loop simulation experiment is built,and the verification based on ADHDP regulator is completed.The experimental results show that the proposed algorithm can effectively improve the motor speed regulation performance.