摘要：伺服系统转动惯量的变化会影响伺服控制性能，为提高伺服性能需要对转动惯量进行在线辨识。在详细介绍两种永磁同步电机转动惯量实时辨识方法的基础上，提出了融合模型参考自适应算法和负载转矩观测器的混合观测模型。该混合观测模型采用模型参考自适应算法在线辨识电机转动惯量，建立降阶负载转矩观测器，实时观测负载转矩和粘性摩擦系数。仿真实验验证了该方法的有效性。
关键词：交流伺服系统；转动惯量辨识；模型参考自适应；负载转矩观测器
Abstract: The inertia variation of the AC servo system highly influences the servo performance. In order to improve the dynamic performance of the servo system, its necessary to identify the system inertia at real time. Based on the analysis of two online inertia identification methods, the hybrid observer model was presented. The MRAS method was used to identify the inertia. The orderreduced load torque observer was obtained to identify the load torque and viscous coefficient. Simulation results show the effectiveness of the proposed methods.
Key words: AC servo system；inertia identification；MRAS；load torque observer