基于MRAS的感应电机无速度传感器矢量控制

Speed Sensorless Vector Control of Induction Motor Based on the Model Reference Adaptive System

在两相静止坐标系下的模型参考自适应系统(MRAS)中,由于转子磁链在α与β轴的两个分量之间存在交叉反馈关系,离散计算时收敛性能较差。采用了一种在两相旋转坐标系下实现的模型参考自适应系统,来辨识感应电机的转速。该方法的参考模型和可调模型分别由改进的电压模型和两相旋转坐标系下转子磁链的电流模型构成。在以TMS320C32为核心的实验平台上进行了实验研究。实验结果表明,该系统能较好地估计电机的磁链及转速,具有良好的稳态辨识特性。

In the model reference adaptive system （MRAS） implemented in two axis stationary reference frame, the convergence performance is poor due to the cross feedback relationship between the a-axis compo nent and β-axis component of the rotor flux when the discrete-time calculation is operated. A MRAS implemented in two axis rotating reference frame was proposed to identify the rotor speed of an induction motor. The reference model and adjustable model used in this MRAS scheme are composed of an advanced rotor flux volt age model and rotor flux current model in two axis rotating reference frame respectively. Experimental studies are carried out on the induction motor experimental platform. TMS320C32 DSP is used as the controller. Ex perimental results show that the system can estimate the flux and speed with good accuracy, and has a better steady identification characteristic.