On-line testing of a fuzzy-neuro based protective relay using a real-time digital simulator

This paper presents a new approach with a fuzzy-neuro based protective relay for real-time fault detection and classification in power transmission systems. The integration of neural network technology into the fuzzy logic controller enables the fuzzy rules to be learned by examples of input/output pattern pairs. From the power system point of view, the symmetrical components in combination with three line currents are utilised to detect fault types such as single line-to ground, line-to-line, double line-to-ground and three line-to-ground, and then to define the faulty line. The operating principle of the fuzzy-neuro protective relay, which depends on currents alone (without having to make reference to the system voltage) is, by conventional definition of distance protection, a novel approach. One major advantage of this technique, therefore, is the potential savings on voltage transformers in the actual implementation. Real-time test results are presented in this paper and they indicate that this approach can be used as an effective tool for high speed digital relaying, as the correct detection is achieved within a cycle of the fault occurrence