An approach for on-line electrical machine winding inter-turn fault detection

On-line machine fault detection is preferable since it is difficult to move and dismantle an electrical machine once it is installed and operated on-site. Hence we propose an on-line stator winding inter-turn fault detection approach based on electromagnetic field signature. Search coil technique will be used to capture the main flux that pass through stator tooth. The magnetic flux distribution is obtained by an electromagnetic field model. This fault detection scheme is suitable for non-stationary operating electrical machine system. A permanent magnet synchronous machine will be used as the candidate machine in this study. Finite element method simulations are used to validate this proposed on-line detection scheme.