On-line fault diagnosis of power systems by a new expert system

This paper proposes a new neural networks diagnostic system for on-line fault diagnosis of power systems using information of relays and circuit breakers. This system has a similar profile to an expert system, but can be constructed much more easily from a few sets of training samples. These samples are used for training of subnets of three neural network modules. The diagnostic system can be applicable to the power system control center for single or multiple faults. Even in the cases of failure operation of relay and breaker, or error-existent data transmission. This system can detect, classify and locate faults. The proposed approach has been practically examined on a model power system with real events. Obtained results confirm the validity of the developed approach.