Electrical Signature Analysis for Condition Monitoring of Permanent Magnet Synchronous Machine

Abstract

Permanent magnet synchronous machines (PMSMs) drives are attractive and have been used in several applications, because of their noteworthy advantages. In some applications, the continuous operation is necessary, and then the PMSM drive outage is unacceptable or may cause great losses. Thus, several studies have been accomplished in order to detect incipient faults in PMSMs. In this context, the electrical signature analysis (ESA) technique is highlighted, because of the feasibility and non-invasive features. ESA allows the fault detection by only analyzing the electrical machine quantities. This paper proposes a study of ESA for PMSM condition monitoring. The review of some fault patterns is presented as well as the development of a scale model laboratory to simulate faults in a real PMSM in operation. The PMSM is used to drive a fluid pumping system, and different types of fluids are tested, being different load conditions from the PMSM point of view. The presented results are promising, encouraging ESA based methodologies for PMSM fault detection.

Keywords

• condition monitoring
• digital signatures
• fault detection
• permanent magnet machines
• predictive maintenance