Stator Fault Detection and Faulty Phase Identification in Network/Inverter-fed Induction Machines using Negative Sequence Current Component

Abstract

This paper proposes a non-invasive negative sequence impedance-based technique to detect stator turn-to-turn faults (STTFs) and dentify the related faulty phase at early stages based on the tracking the magnitude and angle variations of the negative sequence current component generated due to STTFs. To extract these indicators, a simplified steady-state negative sequence equivalent circuit of the induction motor is used. To neutralize the effect of various produced disturbances by the inherent non-ideal construction of the machine and also unbalanced feed voltage to the STTF diagnosis, they will be estimated and removed from the main obtained component. It is shown experimentally that the introduced technique is independent of mechanical loading level (load variations) and is applicable for network or inverter-fed motors as well. Online fault detection and faulty phase identification, as the most important goals of the protection plan, are accessible by defining an appropriate threshold for the magnitude and allowable range of angle variation of the introduced criterion, respectively. The performance of the method is evaluated by simulation as well as multiple experimental tests. The experimental results have shown that from the sensitivity point of view, even weak faults are detectable by such a technique. Also, the obtained tests showed that such technique is robust, reliable and secure in the face of unbalanced voltage sources and load level variations. In addition, the performance of this method for the inverter-fed mode showed that the related sensitivity will be increased in such a condition.