Non-Invasive Fault Detection of Stator Windings of Induction Motors

Abstract: Condition monitoring of induction motor has been widely researched over recent years due to its ability to monitor operating characteristics and the health status of induction motor. Various methods have been used to monitor induction motors such as thermal monitoring and vibration analysis. This paper introduces an alternative method which is to use an inductive coupling method to extract in-circuit impedance of induction motor. This method allows for an online measurement of the system under test (SUT) preventing any unnecessary shutdowns. These unnecessary shutdowns may incur a loss of revenue for relevant industries that depends heavily on induction motor operations. Two sets of experiments were conducted in this paper, the first experiment was to examine the accuracy of the proposed method by simulating various SUT using simulated resistor, inductor, and capacitor (RLC) network. The next experiment was to detect incipient stator faults such as turn to turn faults in an induction motor. This proposed method measures the impedance of the stator winding of an induction motor to identify abnormalities. In addition, this method allows for an accurate in-circuit impedance measurement without the influence of motor load and frequency changes as well as faults such as bearing and rotor faults.