Development of Artificial Neural Network and Adaptive Neuro-Fuzzy Inference System Based Techniques and Algorithms for Protection of Transmission Line

Abstract

This paper presents a relaying algorithm Based on Artificial Neural Network (ANN), and Adaptive Neuro-Fuzzy Inference System (ANFIS) techniques for the protection of transmission lines. A feed-forward ANN with six inputs and eleven outputs has been developed for the detection and classification of faults. Data has been generated by simulating a 400kV, 50Hz, 100 km transmission line in PSCAD/EMTDC at a sampling frequency of 2kHz. ANN has been found with an accuracy of 100% for fault detection and classification in both training and testing phases with the relay operating time of 12.5ms. ANN has been further trained and tested using full data. Two-fold cross-verification was carried out. An accuracy of 100% was obtained on testing with a 12.5ms delay each time. In addition, the adaptive neuro-fuzzy Inference system with the same inputs and outputs with the ANN has been developed for the detection and classification of faults. The optimum number of epochs for both testing and training was found to be 5 with a training error of 0.0038754, and a testing error of 0.081. ANFIS has the least error, high accuracy, least number of epoch and faster than ANN.