Research On Parameter Identification Of Permanent Magnet Synchronous Motor

Permanent magnet synchronous motor (PMSM) is now widely used in control system like high performance servo systems, aerospace, and wind power generators due to its characteristics of high power density, high efficiency and good control performance. In motor control systems, precise motor parameter values are necessary in the design of the controller. However, the motor stator resistance and inductance parameters changes with the temperature rising and flux saturation. Especially, the impacts on PMSM parameter values from the temperature are the most significant and the most frequently. So the methods for PMSM parameters online identification are deserved to research in the motor control field. This paper analyzes the mathematical model and basic control strategy of PMSM, focusing on the motor parameter identification methods.Firstly, this paper introduces the structure of PMSM and the modern parameter identification technology based on vector control. Then the three coordinate systems related to vector control and the vector transformations are derived. Besides, the mathematical models of PMSM in ABC static coordinate system and d-q rotating coordinate system and the principle and digital implementation of vector control and SVPWM are described in detail.Secondly, the methods for parameter measurement of PMSM are given in this paper, and the effective measurement data are provided to the parameter identification. Then for parameters cannot be changed in existing PMSM models in MATLAB, the PMSM models are created according to the mathematical models in d-q rotating coordinate system. After given the input voltage and the load torque, the current, the voltage and speed of PMSM can be obtained by this model. The simulation results show the validity for the model of PMSM.Thirdly, this paper introduces the parameter identification methods of PMSM in detail. The objects of parameter identification are stator resistance and inductances in d and q axis which change a lot with temperature. For the disadvantages like data saturation and lower real time performance of RLS, the method of least square with restricted memory (RMLS) is proposed. And due to the complexity of calculating the inverse matrix by EKF, a novel improved EKF strategy is proposed to avoid calculation of the inverse matrix. After the theoretical analysis, comparisons of simulation results by the four methods for parameter identification verify the feasibility of the improved algorithm.Finally, the paper briefly described the hardware and software design of PMSM control system on the platform TMS320F2812.