中文摘要：

在电机磁链的两种估计模型中，电压模型相对于电流模型具有所需电机参数少、实现方法简单等优点，从而在电机控制系统中获得了广泛的应用。但电压模型中的纯积分环节存在直流偏置和初始相位问题。该文研究了一种新的电机磁链辨识算法，它是由1个高通滤波器和1个坐标变换环节构成。高通滤波器的截止频率等于电机的同步角频率。该算法可以有效地消除纯积分环节存在的问题，同时不会在辨识的磁链中引入幅值和相位误差，保证电机在稳态运行和转速突变情况下磁链辩识的准确性。该算法结构简单，易于工程实现。该文对提出的新辨识算法进行了仿真，并针对1台异步电机，利用数字信号处理器TMS320F2812DSP对其进行了数字化实现。仿真和实验结果验证了所提出方法的有效性。

英文摘要：

In the two types of motor flux estimation models, the voltage model is widely applied in the motor control system areas because compared with the current type, it has more advantages such as less parameters and easy to realize. But the pure integrator of the voltage model has some disadvantages like DC bias and initial phase problems. A novel arithmetic of motor flux estimation was proposed in this paper. It was made up of a high-pass filter and a coordinate transformation. The cut-off frequencies of the high-pass filter are set to the synchronous frequency of induction motor. This arithmetic can effectively eliminate the direct current bias and original phase problem without importing range and phase error in the flux, so the motor flux including its amplitude and phase can be estimated accurately on stable state and speed step state. Besides, this arithmetic is very simple in structure and easy to implement in industrial application. The experiment is implemented on the induction motor using the digital single processor TMS320F2812DSE Simulation and experiment results have verified the effectiveness of the new arithmetic.