Finites elements and signal processing analysis of eccentricity and demagnetization faults in PMSM drivetrains: Approach for diagnosis

TL;DR

This paper presents a finite element and signal processing approach to diagnose eccentricity and demagnetization faults in PMSM drives, enhancing fault detection accuracy for automotive electric machines.

Contribution

It introduces an analytical-numerical FEMM-MATLAB coupled method for detecting specific faults in PMSMs, combining harmonic analysis and time-frequency techniques.

Findings

Distinct spectral signatures for faults identified

Effective fault detection compared to healthy motor signals

Enhanced diagnostic accuracy for automotive PMSMs

Abstract

Today, interest in automotive applications notably Hybrid Electric Vehicles (HEV) has risen due to environmental concerns and the modern society's energetic dependence. Consequently, it is necessary to study and implement in these vehicle fault diagnosis systems which allow them to be more reliable and safe enhancing its sustainability. Several research works have been focused on Fault Diagnosis approaches for electric machines due to it's influence in the automotive system availability. Our work fits into this context, firstly with a review of faults and diagnosis methods of the permanent magnet synchronous motor (PMSM) and secondly by using analytical-numerical approach based on finite element method and signal processing (FEMM-MATLAB coupling) in order to detect and analysis of eccentricity and partial demagnetization faults for surface mounted PMSM. So, this work mainly includes fault performance, harmonic component and signal spectrum characteristics, timefrequency analysis technique. The outcomes obtained are further compared with the motor characteristics under healthy operation to have distinct detection to faults.