Study On Identification Of Permanent Magnet Synchronous Motor

With the development of the times,because PMSM have high operating power,low maintenance cost,a faster response speed and sturdy struct advantage,et al.It can be applied to various civil electrical appliances,transportation vehicles,elevators,transportation machinery and other fields.It is easily affected by temperature,magnetic saturation and other environmental factors in the operation of PMSM.Its internal parameters are dynamically changing resulting in reliability of the entire system decreases and performance of the control system is impaired during normal operation.Therefore,it is important to identify the parameters of PMSM accurately and quickly.Aiming at the problem that the particle swarm algorithm cannot identify the parameters of the PMSM quickly and accurately because of the reason that particles will being precocious.This paper proposes a self-adaptive chaos particle swarm optimization algorithm,which is used to determine the parameters of permanent magnet synchronous motors.After analyze the convergence ability of the algorithm in the standard test functions such as singlepeak and double-peak.This algorithm is used to determine the electrical parameters of PMSM.Based on MATLAB/SIMULINK for simulation research,The simulation results show that the adaptive chaotic particle swarm optimization algorithm is more accurate and fast than the standard particle swarm optimization algorithm in the off-line parameter identification of PMSM.Aiming at the problem that the recursive least square method of PMSM can not identify the parameters on line quickly because the recursive least square method appear data flooding.This paper introduces a recursive least square method with forgotting factor and realized its application in the on-line parameter identification.The simulation results show that the algorithm converges faster.At the same time,the parameters of PMSM are identified online by changing the forgetting factor.The simulation results show that the forgetting factor will affect the convergence and stability of the identification.