SOFT-COMPUTING BASED DATA MINING

In this thesis a noble method of fault locating in a radial power distribution system is discussed. Interval type-2 fuzzy system will ease the calculation involved. The history associated with fault occurred in power system can be effectively used as a strong database. This database will act as the trainer to the interval type-2 fuzzy expert system. A rich history of faulty area will enhance the performance of the interval type-2 Fuzzy Fault Identification System (IT2FFIS). Fault section location aims to identify the faulty area in the power system using the post fault status of protective relays and circuit breakers and diagnose the cause of a fault that has occurred. Restoration cost can be significantly reduced by accurately identifying a fault location and its cause, which also reduces the restoration time of the faulty section. In this thesis methodology based on interval type-2 fuzzy expert system is discussed for assistance to the operators for easy fault detection.The interval type-2 fuzzy sets are used here for data mining for fault location in the distribution system.