Detection of winding faults in auxiliary and main windings of single phase squirrel cage induction motor by monitoring Lissajous patterns

The artificial neural network (ANN) is mostly reported in the literature for the detection of winding fault in single phase squirrel cage induction motor (SPSCIM). This technique required expert opinion and rigorous training. In order to overcome this problem, the Lissajous Pattern (LP) based technique is proposed. In this paper, the Lissajous patterns of combination of supply voltage, current in main winding and current in auxiliary winding are used for the detection and classification of winding faults. Under normal running condition, these Lissajous patterns are not circular in shape. In order to have circular shape, a correction factor is added in supply voltage, current in main winding and current in auxiliary winding. During winding faulty condition the amplitude and phase angle of main and/or auxiliary winding changes, which further lead to change in shape, size and orientation of Lissajous patterns. By monitoring shape, size and orientation of Lissajous pattern, the type of winding fault and its severity with location is identified. A 3 hp capacitor start-capacitor run single phase squirrel cage induction motor is used to verify the proposed technique.