A Multifaults Online Detection and Identification Method for Concentrated Winding PMSM Using Search Coil Array

Concentrated winding permanent magnet synchronous motors (CW-PMSMs) are susceptible to inter-turn short-circuit fault (ITSCF), uniform demagnetization fault (UDF), partial demagnetization fault (PDF), and combined faults during their operational lifespan. The intricate and interconnected nature of these single and composite faults presents significant challenges in their unified detection and isolation, resulting in a lack of comprehensive research on collaborative diagnosis for multiple faults. Addressing this challenge, this study presents an approach for the collaborative detection and identification of multiple faults in CW-PMSMs, employing a search coil array (SCA). This method not only enables the early detection of the faults mentioned above but also facilitates the determining of the specific fault scenario upon detection, utilizing a 4-D fault feature vector based on the voltage signal of the SCA. By strategically placing multiple toroidal coils within the stator slot, the SCA captures information from fault locations associated with these faults, which are critical for early fault diagnosis. Additionally, in the presence of the faults mentioned above, the voltage waveforms from the SCA exhibit distinctive characteristics, thereby effectively identifying these faults. To ensure the reliability and accuracy of this fault diagnosis method under varying factors like fault parameters and operating conditions, this study formulates four fault indicators based on the SCA's voltage signal. These indicators effectively differentiate these faults while simplifying the diagnostic algorithm. Simulation and experimental results verify the efficacy of this method for online detection and identification of these single and composite faults in all CW-PMSMs under steady-state conditions. Early detection and accurate identification of these faults can not only prevent further deterioration of faults but also provide important information for fault-tolerant control and maintenance of the motor. This research introduces an innovative approach to motor fault diagnosis and forms a crucial basis for fault-tolerant control in motors.