Comprehensive study of mixed eccentricity fault diagnosis in induction motors using signature analysis

► Reviews various signature analysis for mixed eccentricity detection. ► Instantaneous power and power factor signature analysis valid for unbalance voltage conditions. ► Eigen value based mixed air gap eccentricity detection. ► More than one eccentricity characteristic frequencies are considered for prediction. ► Prediction is based on relative increase in eccentricity from installation stage. Modeling and simulation studies of an induction motor always help in identifying the parameter to characterize the asymmetrical fault in the machine. Hence in this paper, an air gap eccentric induction motor is modeled using multiple coupled circuit approach and 2D-Modified Winding Function Theory. The machine model is simulated under different eccentricity conditions to obtain the motor current spectra, power spectra and power factor spectra to detect the eccentricity related frequency components and the results are compared. All these analysis are based on the variation in the amplitude of mixed eccentricity related frequency component in these parameters with the variation in the level of eccentricity in the machine. A new fault severity detection method based on co-variance analysis is presented to predict the degree of deterioration in the health of the machine due to air gap eccentricity from the installation stage.