Fuzzy Logic Based Fault Type Identification in the Radial LT Power Distribution Feeder

Fault classification is necessary for the rapid restoration of service to LT consumers after the occurrence of a fault. This paper presents a step by step procedure for the identification of ten different types of faults commonly occurring in the LT distribution system. Information on the distribution transformer secondary current for different faults at different load buses is used to define the input fuzzy variables. Fuzzy inference engine and the centroid de-fuzzifier are used to relate the input to the fuzzy rule base and to obtain crisp outputs respectively.