Electromagnetic flux-based condition monitoring for electrical machines

The main aim of this paper is to study the ability of various electromagnetic fluxes to enhance the detection and localization accuracy of faults in a 35 kW cage-induction motor. Another aim of this paper is to study the modifications brought by various stators winding design to some of the asymmetrical air-gap electromagnetic flux density harmonics responsible for the detection of different faults. For this purpose, the following designs are considered: stator winding formed of no parallel branches, 2 and 4 parallel branches respectively. For the design consisting of no parallel branches we have considered the stator winding formed of one and two layers respectively. The studied faults consist of inter-turn short circuit in the stator winding, rotor-cage related fault (bars and end-ring breakage), eccentricities (static and mixed) and bearing failure. The relevant fault signatures of the electromagnetic fluxes are issued both from measurements and from two-dimensional numerical electromagnetic field simulations at steady state. When possible, the experimental verification for the ldquovirtualrdquo measurement signals provided by the numerical electromagnetic field simulations is achieved.