Fault Classification in Transmission Lines Using Radial Basis Function Neural Networks

Fault classification in transmission lines is essential for maintaining a stable, efficient, and safe power supply, which is fundamental to modern society’s functioning and economic well-being. The wavelet transform-based techniques have limited performance for fault detection and classification in transmission lines since they rely on threshold limits. Since, the core issue is that the optimal threshold value depends on the specific power system network, making it difficult to establish a universal threshold for accurate fault classification. This research article proposes the use of a radial basis function neural network (RBFNN) with wavelet transform for fault detection and classification in transmission lines. This study investigates the challenge of threshold selection in wavelet transform-based fault detection techniques, highlighting that RBFNNs do not require pre-defined thresholds for fault identification. The paper also emphasizes the importance of extensive training and testing data to achieve high accuracy. This approach offers a promising alternative to traditional wavelet transform methods by eliminating the need for network-specific threshold selection.