Feature Extraction of Demagnetization Faults in Permanent-Magnet Synchronous Motors Based on Box-Counting Fractal Dimension

This paper presents a methodology for feature extraction of a new fault indicator focused on detecting demagnetization faults in a surface-mounted permanent-magnet synchronous motors operating under nonstationary conditions. Preprocessing of transient-current signals is performed by applying Choi-Williams distribution to highlight the salient features of this demagnetization fault. In this paper, fractal dimension calculation based on the computation of the box-counting method is performed to extract the optimal features for diagnosis purposes. It must be noted that the applied feature-extraction process is autotuned, so it does not depend on the severity of the fault and is applicable to a wide range of operating conditions of the motor. The performance of the proposed system is validated experimentally. According to the obtained results, the proposed methodology is reliable and feasible for diagnosing demagnetization faults in industrial applications.