Rotor fault diagnosis in asynchronous machines via analysis of the start-up transient into intrinsic mode functions

A novel approach for rotor fault diagnosis during the start-up process is proposed. Rotor faults create asymmetries that affect the phase currents in a subtle yet distinctive way. The approach presented here is based on the application of Empirical Mode Decomposition (EMD) to the measured start-up current for the extraction of a specific complex Intrinsic Mode Function (IMF). The characteristics of this particular IMF are quite distinct for the case of an asynchronous machine with rotor asymmetry and can be used to discriminate between faulty and healthy machines. This task of fault diagnosis is performed via the use of two Hidden Markov Models (HMMs) which are well known for their temporal pattern recognition capabilities. Simulation and experimental data demonstrate the effectiveness of the proposed methodology for the detection of rotor asymmetries.