Characteristic signature identification of air-gap eccentricity faults using extended d-q model for three phase induction motor

Abstract

The supremacy of three phase squirrel cage induction motors in industrial drives demands accurate and reliable diagnostics for condition monitoring and internal fault detections. Operating stresses on these machines are electrical, mechanical, thermal, magnetic and environmental in nature and might result in internal faults. Avoiding unscheduled maintenance and repair intervention can prevent losses in money, material, manpower and time in process industries. Detection of faults in its early stage becomes an indispensable need especially in critical applications. Mathematical model based simulation studies will support fault signature identification to a great extent. Conventional d-q model of AC machines are not generally used for internal fault diagnoses. In this paper a novel attempt is made for simulating eccentricity related faults by modifying conventional d-q model of three phase induction motor. Characteristic fault signatures were identified in the stator current frequency spectrum for static, dynamic and mixed eccentricity conditions. The increase in magnitudes of these characteristic frequency components with increase in severity of faults is also established through model based simulation studies. The experimental study results presented for static eccentricity in a three phase squirrel cage induction motor clearly validates the modelling approach.