The abnormal torque recognition of DC motor via an efficient novel algorithm for the chip embedded

In this paper, a novel method of the signal sampling integration combined with the signal edge detection is proposed to inspect the normal and abnormal loads of the DC brushless motor. From the experimental results, the designed torque observer not only can obtain the normal load events but also is more suitable for the abnormal motor load detection. In addition, the proposed scheme of signal integration could identify the noise interference as well as enhance the accuracy of fault detection by adjusting the sampling integration period. On the other hand, the experimental results are also verified by the Wavelet analysis. Because the computational load and the data processing via this efficient approach are fewer than the Wavelet transforms for the abnormal load detection, the algorithm of the designed motor torque observer is more explicit and easier to implement on a digital chip for the motor driver.