Diagnosis of Induction Machines' Rotor Faults in Time-Varying Conditions

Abstract

Motor current signature analysis is the reference method for the diagnosis of induction machines' rotor faults; however, in time-varying conditions, it fails as slip and speed vary, and, thus, sideband components are spread in a bandwidth that is proportional to the variation. Variable speed drive applications are common in the aerospace, appliance, railway, and automotive industries and also in electric generators for wind turbines. In this paper, a simple and effective method is presented that allows the diagnosis of rotor faults for induction machine drives in time-varying conditions. It is tailored to direct rotor flux field-oriented controlled drives, where the control system provides suitable signals that are exploited for the demodulation to a constant frequency of time-varying signatures related to the rotor faults. Simulations and experiments are reported to validate the proposed method on a critical speed transient.