基于人工鱼群算法的PMSM参数辨识

PMSM parameter identification based on artificial fish swarm algorithm

为了提高永磁同步电动机(Permanent-Magnet Synchronous Motor,PMSM)的参数辨识度,提出一种基于人工鱼群算法(Artificial Fish Swarm,AFS)的PMSM参数辨识方法。以PMSM待辨识参数作为位置向量,以d轴、q轴电流误差的平方和与时间乘积的积分作为优化目标,给定待辨识参数范围,通过迭代寻优得到待辨识参数的最优值。仿真结果表明:相比于粒子群算法,AFS算法在辨识PMSM电阻时,误差降低了3.11%,辨识d轴和q轴电感时,误差分别降低了2.45%和1.60%,辨识永磁磁链时,误差降低了3%,具有更高的辩识度。

In order to improve the accuracy of parameter identification of the Permanent Magnet Synchronous Motor(PMSM),a parameter identification method based on Artificial Fish Swarm(AFS)algorithm is proposed.This method uses the PMSM parameters to be identified as the position vector.Then the integration for time multiplied square of d,q axis current error(ITSE)is used as the cost function.Finally,the optimal values of the parameters will be obtained through iterative optimization when these parameters range are given.The simulation results show that the AFS algorithm has higher accuracy comparing with the particle swarm algorithm,because when uses the AFS algorithm to identify the parameter of the PMSM the resistance error is reduced by 3.105%,the d,q axis inductance errors are reduced by 2.45% and 1.60% respectively,the permanent magnet link error is reduced by 3%.