A Review of Separating Mechanical Load Effects from Rotor Faults Detection in Induction Motors

Abstract

Motor current signature analysis (MCSA) is usually used to detect rotor faults in induction motors. However, some abnormal mechanical load conditions such as load torque oscillation have similar effects in the stator current spectrum and often cause ambiguity in online motor condition monitoring. This paper first compares the effects of different types of rotor faults and mechanical load anomalies on the stator current spectra. After that, a comprehensive literature review is presented to evaluate different diagnostic techniques on separating these two effects. Based on their respective merits and limitations, an effective and computationally efficient strategy is proposed to reliably detect rotor faults under arbitrary load conditions without dependence on the accurate estimation of motor parameters.