An on-line knowledge-based system for fault section diagnosis in control centers

This paper presents a knowledge-based system for online diagnosis of power system fault allocation (SIDUF-TR), a decision making support for the energy control center operators in Mexico. A new inference method for possible faulted elements is proposed, which uses the real-time information about tripped relays and circuit breakers to generate a decision tree describing the operation of the protection schemes in the power system. The inference method conforms the branches of decision trees performing a classification process which identifies the type of relay and circuit breaker operation for different possible faulted elements. Architecture for real-time operation of SIDUF-TR and the result of its application to a real disturbance are also presented in this paper.