Empirical mode decomposition and Hilbert-Huang transform-based eccentricity fault detection and classification with demagnetization in 120 kW interior permanent magnet synchronous motors

Permanent magnet synchronous motors (PMSMs) are extensively used in electric vehicles (EVs) owing to their high power density and efficiency. However, their safety and reliability have not been verified sufficiently. Failure of PMSMs in EVs can cause performance degradation, increased maintenance costs, and catastrophic accidents. Therefore, early detection of faults in PMSMs is necessary for an efficient and safe operation. In this study, static, dynamic, and mixed eccentricity faults in 120 kW interior permanent magnet synchronous motors (IPMSMs) were diagnosed and distinguished from uniform demagnetization fault by monitoring the Hilbert spectrum of the stator current. Experiments were carried out for different speeds, loads, and fault severities to analyze the effect of operating conditions on the proposed fault detection method. The phase current data were decomposed into set of intrinsic mode functions (IMFs) with empirical mode decomposition (EMD) method and Hilbert transform was applied for each IMFs to calculate the instantaneous frequency and energy. The resultant Hilbert spectrum exhibited instantaneous frequency distortions with an instantaneous energy surge under an eccentricity fault. The effect of thresholding in EMD was analyzed to improve the accuracy of the proposed method under high-load and high-speed conditions. Also, the experimental results showed that eccentricity and uniform demagnetization faults can be classified by monitoring the overall instantaneous energy.

Automated summary

In this study, faults in permanent magnet synchronous motors (PMSMs) were diagnosed and analyzed by monitoring the Hilbert spectrum of the stator current. Experimental results showed that fault detection can be improved by monitoring the overall instantaneous energy.