Detection of static eccentricity and turn-to-turn short circuit faults in permanent magnet synchronous AC machines

Permanent Magnet Synchronous Machines play a major role in many industrial applications, therefore developing a method to detect the faults, fault type and severity becomes more important. Faults will cause changes to the machine performance and parameters. One of these effects is changing the machine saturation level. This change is reflected in the incremental inductance curve as a change in the peak amplitude and/or a shift in the curve peak. In this paper the difference in the incremental inductance curves between healthy and faulted machine is used as a fault indicator. Based on the direction and the amount of the curve peak shift, the type of the fault can be detected and its severity can be estimated. Two faults are discussed here: static eccentricity and turn-to-turn short circuit. k-Nearest Neighbor classifier is used to detect the type and estimate the severity of the fault. Both simulation and experimental data are used to validate the method. Finite Element Analysis simulations using Flux-2D software are performed on two 3 phase Permanent Magnet Synchronous Machines with different winding topologies: a distributed and a concentrated winding machine, experimental data are performed on the distributed winding machine.