ClasLoc : A Robust Fault Management Methodology for Transmission Lines Using RBFNN and ANFIS

Abstract

Transmission lines are crucial for efficient and reliable electricity delivery of electricity but can also be susceptible to faults due to various factors such as external interference, device failures, and ambient conditions. The reliability, accuracy, and efficiency of fault management on the power grid depend heavily on fault classification and fault location. This study integrates a fault detection and classification system for transmission lines and an automatic fault localisation system (ANFIS) for fault location to present a comprehensive approach to managing transmission line problems. The proposed methodology aims to address the technological challenges associated with fault management in power grids by leveraging existing research on radial basis function neural networks (RBFNN) and ANFIS. The integration and validation of the proposed methodology involve incorporating improved fault classification via RBFNN and robust fault locating through ANIS. Simulations and real-world testing will validate the integrated methodology, assessing its performance in various fault scenarios and system configurations. The WSCC 9-bus system is used to validate and test the proposed design, which includes power plants, transformers, transmission networks, and distribution networks. The ClasLoc system has a classification accuracy of 90% and a location accuracy of 89%, ensuring a safe working environment.