永磁同步电机五参数同时辨识方法研究

Research on Five-parameters Identification Method of Permanent Magnet Synchronous Motor Based on Bacterial Chemotaxis Particle Swarm Optimization

精确获取永磁同步电机(PMSM)参数是使其获取高性能驱动控制基础。针对PMSM多参数辨识较为困难的问题,本文提出基于细菌趋化粒子群算法(PSOBC)的PMSM五参数辨识方法,该辨识方法可同时辨识定子绕组电阻,dq轴电感,永磁体磁链,转动惯量五个参数。在PMSM矢量控制id=0基础上,采取在定子d轴注入负序弱磁方波电流策略,并在两种控制策略条件下推导出含五个待辨识参数的满秩离散数学方程组。PSOBC算法在基本粒子群算法基础上,加入细菌排斥策略,能有效维持迭代后期粒子种群的多样性,从而提高其寻优性能。经实验验证了此辨识方法的有效性和精确性。

Accurate acquisition of permanent magnet synchronous motor(PMSM)parameters is the basis for high performance drive control.Aiming at the difficult problem of PMSM multi-parameter identification,this paper proposed a PMSM five-parameter identification method based on bacterial chemotactic particle swarm optimization(PSOBC),which can identify stator winding resistance,dq-axis inductance,permanent magnet flux linkage,and moment of inertia,five parameters.Based on the PMSM vector control id=0,the negative sequence weak magnetic square wave current strategy was injected into the stator d-axis,and the full rank discrete mathematical equations with five parameters to be identified are derived under two control strategies.The PSOBC algorithm added a bacterial rejection strategy based on the basic particle swarm optimization algorithm,which can effectively maintain the diversity of particle populations in the late iterative stage and improve its optimization performance.The validity and accuracy of this identification method were verified by experiments.