A New Internal Fault Detection and Classification Technique for Synchronous Generator

A new internal fault detection and classification technique for synchronous generators is presented. First, the presence of an abnormal/fault condition is detected based on the Negative Sequence Component (NSC) of neutral end current. Thereafter, the internal fault and abnormal/external-fault conditions are discriminated by utilizing the instantaneous phase angle between the NSCs of the terminal voltage and current. Subsequently, the internal fault is classified based on phase angle between NSCs of terminal voltage and neutral end current along with the zero sequence component of the neutral end current. Efficacy of the proposed scheme has been verified on a Phase Domain Synchronous Machine model developed on the Real Time Digital Simulator (RTDS̅). The results derived from various fault data sets show that the proposed technique is able to effectively distinguish between internal and external faults including special cases such as current transformer saturation, overloading and unbalance condition. Further, it provides added capability of detecting and identifying inter-turn faults irrespective of the winding configuration of the generator. Comparative assessment of the proposed scheme in terms of detection time and incorporation of different types of faults with the existing techniques proves its superiority.