Detection of rotor eccentricity faults in closed-loop drive-connected induction motors using an artificial neural network

This paper focuses on the detection of mixed air gap eccentricity in a closed-loop drive-connected induction motor. It analyzes the distribution of fault signatures, and proposes a diagnostic scheme that monitors both the stator voltage and current space vectors. Since these space vectors are readily available control variables, this detection method does not add any extra cost to the existing instrumentation system of the motor drive. For a drive-connected induction motor, its speed varies widely. The amplitudes of eccentricity related components change non-monotonically with the operating conditions. An artificial neural network (ANN) is used to learn the complicated relationship and estimate corresponding signature amplitudes over a wide range of operating conditions. Experimental results from a three-phase induction motor driven by a commercial vector-controlled drive validate the feasibility of this diagnostic scheme.