基于参数辨识的内置式永磁同步电机最大转矩电流比电流预测控制

Maximum torque per ampere current predictive control of interior permanent magnet synchronous motors based on parameter identification

为了实现对内置式永磁同步电机的高效准确控制,解决电机运行过程中参数变化对控制性能的影响,提出了一种基于在线参数辨识的内置式永磁同步电机最大转矩电流比电流预测控制方法。根据内置式永磁同步电机的转矩特性对最大转矩电流比的d-q电流最优关系进行化简以便于工程计算;并针对电机参数变化对最大转矩电流比工作点偏移的影响进行了分析。同时,对MTPA算法影响较大的关键参数q轴定子电感,转子永磁体磁链进行基于参考模型的自适应辨识,以确保实时的最优d,q轴电流分配。在得到准确辨识的参数和最优电流指令的基础上进行电流的鲁棒预测控制,使实际电流更快地跟踪指令,提高系统的动态性能。实验结果表明q轴定子电感,转子永磁体磁链在线辨识的误差分别小于3%,3.5%,收敛时间小于20 ms;电机能有效跟踪最大转矩电流比工作点,电流响应时间小于30 ms,能够满足内置式永磁同步电机系统稳定可靠、高效快速等运行要求。

To realize an efficient and accurate control of Interior Permanent Magnet Synchronous Motors(IPMSMs)and solve the effects of the motor parameter changes on the control performance,a method for the Maximum Torque Per Ampere(MTPA)current predictive control of the IPMSM based on online parameter identification was proposed.First,according to the torque characteristic of the IPMSM,an optimal relationship of d-and q-axis currents under the MTPA control was simplified to facilitate engineering calculation.The effects of the motor parameters on the MTPA operating point offset were analyzed.In addition,key parameters q-axis inductance and permanent magnet linkage of rotator,which significantly affect the MTPA algorithm,were adaptively identified based on a reference model to calculate the optimal d-q current distribution in real time.Subsequently,based on the accurately identified parameters and optimal current commands,predictive current control was applied such that the actual current can track the command faster and improve the dynamic performance of the system.The experimental results show that the errors of the online identification of q-axis inductance and permanent magnet linkage of rotator are less than 3%and 3.5%,respectively,and the convergence time is less than 20 ms.The motor can effectively track the MTPA operating point,and the current response time is less than 30 ms,which satisfies the requirements of stable,reliable,efficient,and fast operation of IPMSM systems.