DETECTION OF GENERATOR STATOR FAULTS. USING NEURO-FUZZY TECHNIQUE

Abstract

In this work, Neuro-Fuzzy Technique has been used for developing an algorithm for protection and monitoring of synchronous generator. The developed scheme consists, of two parts; these are (1) Simulation of various operating states of synchronous generator and (2) Fault detection using neuro-fuzzy technique. In first part various operating states, namely, normal operating state, internal fault and external fault state of synchronous generator have been simulated. Modelling of generator has been done in terms of phase quantities. The fault detection part uses current samples from the line-side and neutral-end in addition to the samples from field current. Neural Network is used for detecting operating state and Fuzzy logic is for final decision making to issue a trip signal. Using this developed software application, various operating states of synchronous generator connected to-infinite ,bus can be simulated and a trip signal can be issued in case of short circuit faults in stator winding