Feasibility study of intelligent fault location estimation methods for double-circuit transmission lines

Fault location estimation methods have been suggested by researchers decades back after the development of transmission lines for power transmission. But there are several drawbacks exist such as incapability to locate high resistance faults, boundary faults, intercircuit faults, change in line parameters, evolving faults, etc. Hence, a method should be developed, which can address all the problems of conventional fault location estimation methods. One of the solutions can be artificial intelligence (AI) due to the ability to learn and think. In this article, an AI-based fault location estimation method has been suggested for transmission lines, which use the measurement from one end of the line. A comparative study of four AI methods has been carried out, and one suitable method has been chosen. Three methods are used for signal processing, and one suitable method has been chosen for signal processing and feature selection. The proposed AI-based method has error in location within 1 km for most fault cases, and maximum error in locating fault is 2.91 km. The performance enhancement of the proposed AI-based method over existing methods is that it does not require a communication link as it uses only one end signal; it can locate high resistance faults, boundary faults, intercircuit faults, unaffected by variation in line parameters, variation in power flow angle, etc. The proposed method can be implemented in real power network, which is feasible, easy, low cost, and robust.