Motor bearings and insulation system condition diagnosis by means of common-mode currents and shaft-ground voltage correlation

Nowadays, motor condition or health-state diagnosis is clearly an important issue in the industrial scope. In this paper, a review on motor bearing currents is offered, and a novel induction motor condition diagnosis methodology is proposed, with the focus on bearings and insulation system condition. The proposed methodology is based on the correlation between instantaneous input currents sum and shaft-ground voltage. It can be also used to other kind of diagnosis, such as to detect significant motor electromagnetic asymmetries. A user-friendly system for acquiring and processing the data using artificial neural-networks for the new diagnosis strategy is described. The proposed method should be intended as a complementary technique to other well-known methods for winding and rotor faults diagnosis using instantaneous line currents and mechanical vibration. This paper should be intended as a kick-off, since only the principles and system setup for the method implementation are addressed. Experimental validation is not provided, although some experimental data is used to demonstrate the relations behind the proposed technique.