Hilbert-Huang Transforms for fault detection and degradation assessment in electrical motors

The present work develops a methodology for the analysis of transient signals for fault detection and diagnosis in electrical motors. The objective of this work is the initial development of a Prognostic and Health Monitoring System (PHMS) for the demagnetization of the magnetic field source of a Permanent Magnet Synchronous Motor (PMSM) used in electrical vehicles. The developed methodology is based on the Hilbert-Huang Transform (HHT), a technique which is particularly suitable for processing oscillating transient signals, such as the stator currents typical of automotive electric traction machines. The HHT represents a time-dependent series in a two-dimensional time-frequency domain by extracting instantaneous frequency components within the signal through an Empirical Mode Decomposition (EMD) process. The developed framework has been applied to four transients simulating different levels of demagnetization of the permanent magnet of the PMSM; the obtained results show that HHT enables us to detect and assess the degradation level for a demagnetized core of a PMSM.