An Artificial Neural Network Approach for Three-Zone Distance Protection

• Published in: International journal of modelling & simulation Vol. 25; no. 4; pp. 291 - 298
• Main Authors: Feilat, E.A., Al-Tallaq, K.N.
• Format: Journal Article
• Language: English
• Published: Anaheim, CA Taylor & Francis 01.01.2005; Calgary, AB Acta Press; Zürich
• Subjects: Applied sciences; digital protective relaying; Distance protection; Disturbances. Regulation. Protection; Electric, optical and optoelectronic circuits; Electrical engineering. Electrical power engineering; Electrical power engineering; Electronics; Exact sciences and technology; Neural networks; Power networks and lines; simulation; Performance evaluation; Waveform; Automatic classification; digital protective relaying; simulation; neural networks; Neural network; Distance relay; Implementation; Logic circuit; Experimental result; Transmission line; Feedforward neural nets; Defect localization; Digital circuit; Power system relaying; Numerical simulation; Digital protection; Artificial intelligence; Digital relay; Distance protection; Defect detection
• ISSN: 0228-6203; 1925-7082
• DOI: 10.1080/02286203.2005.11442342

This paper presents a neural network-based approach for three-zone distance protection of transmission lines. The proposed neural-network distance relay comprises three modular identical feedforward neural networks. The first unit is trained for main protection of a transmission line section, whereas the other two units are trained to provide back-up protection for the adjacent line sections. Each neural network is fed with the fundamental frequency voltage and current magnitudes. The output of each neural unit (1 or 0) thus determines if a fault is internal or external to its protection zone. Coordination of the three neural networks and initiation of the tripping signal is implemented through a digital logic circuit. The simulation results presented in this paper show that the proposed neural-network distance relay is very effective in detection and classification of the fault location, and therefore can be considered as a good tool for main and backup digital distance protection.