Current Based Mechanical Fault Detection in Induction Motors through Maximum Likelihood Estimation

This paper proposes a new detection method for induction motor mechanical faults in steady state based on parameter estimation of the stator current. The considered mechanical faults cause periodic load torque oscillations leading to a sinusoidal phase modulation of the stator current. The modulation index is related to the fault severity and can be used as fault indicator. Based on a simplified stator current signal model, the maximum likelihood estimator for a monocomponent signal with sinusoidal phase modulation is derived. The algorithm is implemented using evolution strategies for optimization. The Cramer-Rao lower bounds are calculated and compared to the estimator performance in simulations. The estimation procedure is studied on experimental stator current signals with load torque oscillations and load unbalance