New Formulation through Artificial Neural Networks in the Diagnosis of Faults in Power Systems: A Modular Approach

Abstract

In this work a new method is proposed for the diagnosis of faults in electric power transmission systems based on neural modularity. This method performs the diagnosis through the assignation of a generic neural module for each type of element conforming the transmission system, whether it be line, bus or transformer. A total of three generic neural modules are designed, one for each type of element. These neural modules are grouped together in repetition according to the element to be diagnosed and taking into account its breakers and relays, both primary and back-up. The most important and novel aspect of this method is that only three neural modules are required, one for each type of element, and they can be called upon for the diagnosis as a function, the moment a change of state is detected in any of the breakers, primary or back-up, relating to the element under diagnosis.