Locating stator winding insulation failure in induction machines under different load conditions

This paper aims to investigate stator winding insulation failure (SWIF) in induction motors. The location of the SWIF is considered as a new factor that affects the diagnosis of the stator winding short circuit faults. In case of small rating IMs, removing the whole stator winding is done during maintenance program. However, this is not economical and reasonable for large rating induction motors whilst it is costly and time consuming. Thus, there is a need for an effective condition monitoring system that can detect the faulty section where SWIF is present. In this paper, finite element method is employed in order to model the SWIF in an accurate and inexpensive manner. This modelling technique considers all material properties, machine geometries, non-linear features of core materials, saturation and rotor bar skewing effects. Turn-to-turn and coil-to-coil short circuit faults have been modelled in different stator slots, precisely. Also, achieving a resolution of two turn-to-turn short circuit fault in any arbitrary slot indicates the advantage of the proposed modelling technique over previous methods. The orientation of current vector locus in q-d plane at steady state indicates the SWIF location. In order to improve the resolution and fault detection accuracy, an enhanced phase lock loop system is also employed. The simulation results confirmed the effectiveness of the proposed diagnostic technique for location of turn-to-turn and coil-to-coil stator winding short circuits under different operating load conditions.