基于定子磁链滑模观测器的无速度传感器感应电机模型预测控制

Model predictive control based on the stator flux sliding-mode observer for speed sensorless asynchronous motor

为了对感应电机进行快速精确的调速,本文把电流模型预测控制做为电机的控制方法。通常电流模型预测算法需要电流和磁链的信息,所以又设计了一种用于估算磁链和电流的定子磁链观测器。观测器一般把定子的电流和磁链当做状态变量,用实际检测电流与估算电流的误差来构造滑模面。并将电机状态开关的信号,反馈到磁链、电流的状态方程中,对磁链、电流的观测进行在线校正,并且由李雅普诺夫函数附加关系式来实现对转子角速度的辨识。MATLAB/Simulink仿真结果表明,基于定子磁链滑模观测器的无速度传感器感应电机模型预测控制系统,对比传统矢量控制,该方法提高了系统的动态响应速度和鲁棒性。

In order to control the induction motor quickly and accurately, current model prediction is used as control algorithm. Current and flux information required for model predictive control algorithm, a stator flux observed based on sliding mode control theory is designed. It takes stator current and flux as state variables, sliding mode surface is constructed by the errors of actual current and observed current, The state switch signal is fed back to the state equation of flux and current, correcting the flux and current observations.The identification of rotor angular velocity is also accomplished by an additional relation of Lyapunov function. The MATLAB/Simulink results show that the model predictive control system of sensorless induction motor based on sliding mode observer improves the dynamic response and robustness of the system.