Frequency spectrum investigation and analytical diagnosis method for turn-to-turn short-circuit insulation failure in stator winding of low voltage induction motor

Low-voltage induction motors are widely used in the factories and are becoming very essential components for the industrial process. Improper diagnosis and condition monitoring of induction motor not only leads to high repair expenses but also cause extraordinary financial losses due to their unexpected downtime. In general, turn-to-turn short-circuit failure of stator winding due to their deterioration or damage of electrical insulation is one of the most feasible failures in motor drive systems. This paper proposes a diagnostic method based on amplitudes of characteristic frequency components of load current as features. These features are extracted by close analysis of frequency spectrum and its dependence on load current variation. This method has the advantage of low cost and short data processing time. Also to enhance the accuracy of diagnosis, a novel diagnostic method is proposed with the aid of Support Vector Machine. It is found that the proposed diagnosis method using the features and Support Vector Machine enables the detection of a slight insulation failure like one turn-to-turn short-circuit with practically accepted accuracy.