Inter-Turn Short-Circuit Fault Detection Approach for Permanent Magnet Synchronous Machines Through Stray Magnetic Field Sensing

Fast online detection of inter-turn short-circuit faults for permanent magnet synchronous machines (PMSMs) is a challenging task. Previous attempts based on phase currents or back EMF monitoring lack the capability to accurately locate the inter-turn short-circuit faults in stator windings. This pap...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE sensors journal 2019-09, Vol.19 (18), p.7884-7895
Hauptverfasser: Liu, Xuyang, Miao, Wenchao, Xu, Qi, Cao, Libing, Liu, Chunhua, Pong, Philip W. T.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Fast online detection of inter-turn short-circuit faults for permanent magnet synchronous machines (PMSMs) is a challenging task. Previous attempts based on phase currents or back EMF monitoring lack the capability to accurately locate the inter-turn short-circuit faults in stator windings. This paper proposed a new approach to online inter-turn short-circuit fault detection for PMSMs by means of sensing the external stray magnetic field outside the motor stator yoke. The stray magnetic field provides the information about the phase currents as well as the status of stator windings and thus, the location of inter-turn short-circuit fault can be detected accordingly. In this paper, the inter-turn short-circuit model and the stray magnetic flux model for PMSMs were theoretically analyzed. Both FEM simulations and experiments have been presented to validate the effectiveness of the proposed approach. A sensor array composed of 24 sensitive tunneling magnetoresistive (TMR) sensor units was employed to measure the stray magnetic field distribution outside the PMSM stator yoke. The experimental results show that both the location and severity of the inter-turn short-circuit fault in PMSMs were effectively detected using the proposed approach.
ISSN:1530-437X
1558-1748
DOI:10.1109/JSEN.2019.2918018