硬件在环系统中电机参数辨识及状态估计研究

摘要

矢量控制电机中的敏感参数转子时间常数的实时辨识和电机状态实时估计是电机高性能运行的保证,首先讨论了转子时间常数参数辨识,其次讨论了滑模观测器并给出了观测器稳定性分析,通过观测器得到感应电机磁链状态估计值；将得到的转子磁链值用于MRAS状态估计,得到实时的电机转速.设计了硬件在环仿真系统验证电机状态估计方法及参数辨识方法；硬件在环试验系统包含感应电机,电气测功器以及实时仿真器等硬件,通过硬件在环实时试验得到了实时的电机状态估计及参数辨识结果,试验结果验证了参数辨识及状态估计方法的有效性和实时性.%In vector-controlled induction motor,the on-line estimation of the sensitive parameter rotor time constant and the on-line identification of motor state is the guarantee of motor high performance.A sliding mode observer for induction motor rotor flux identification was studied and the stability analysis of the observer was presented.Then the identified rotor flux was used in MRAS state identification method.A hardware-in-the-loop real-time simulation setup was presented to verify the state identification method and the parameter estimation method.The setup includes an induction motor,a dynamometer and a FPGA-based real-time simulator as real hardware.The experimental results indicate the real-time capability and effectiveness of the MRAS motor state identification and parameter estimation method.