基于遗传算法的车用永磁同步电机在线热模型参数辨识方法  

Parameter Identification Method of Onboard Automotive PMSM Thermal Circuit ModelUsing Genetic Algorithm

作　　者：孙坤迪 董腾辉 张希[1] 朱翀[1] 鲁岩松 SUN Kundi;DONG Tenghui;ZHANG Xi;ZHU Chong;LU Yansong(National Engineering Laboratory of Automotive Electronic Control Technology,Shanghai Jiao Tong University, Shanghai 200240, China)

机构地区：[1]上海交通大学汽车电子控制技术国家工程实验室,上海200240

出　　处：《微电机》2021年第6期55-59,64,共6页Micromotors

摘　　要：随着车用永磁同步电机(PMSM)转速的不断提升,实现对电机绕组与转子温度实时监测在电机热管理系统中至关重要。利用结合电机热模型的在线温升系统同步运算,可以较为经济和准确地对电机关键且难测区域温度进行实时估计。然而,复杂热模型虽然可以保证温度估计的准确度,但会显著提升在线系统的运算量。为解决上述问题,提出了一种新的结合解析化热模型与数据驱动的方法,在实现复杂热模型有效降阶的同时,保证模型精度。经验证,该方法可以将热模型节点数降低超过80%,并使关键区域温度的稳态估计精度达到1.78%,瞬态估计精度达4.18℃,满足热管理系统的快速响应需求和精度要求。With the increasing speed of permanent magnet synchronous motor(PMSM),it is very important to realize real-time temperature monitoring of stator winding and rotor magnets in the thermal management system.The real-time estimation of the temperature in the critical and unpredictable region of the motor can be achieved economically and accurately by using the real-time calculation of the onboard thermal circuit model of the PMSM.However,although the complex thermal circuit model can ensure the accuracy of temperature estimation,it can significantly increase the computational complexity of the onboard system.In order to solve the problems,a new method combining analytical thermal circuit model and data-driven method was proposed,which can effectively reduce the order of complex thermal circuit model and ensure the precision.It was verified that the proposed method can reduce the number of nodes in the thermal path model by over 80%,while ensuring the steady-state and transient estimation accuracy of temperature in key areas around 1.78%and 4.18℃respectively.These accuracies can meet the requirements of the thermal management system.

关 键 词：永磁同步电机 热模型 参数辨识 温升估计 遗传算法 

分 类 号：TM351[电气工程—电机] TM341