Enhanced Fault Detection in Transmission Line Segments using Wavelet Transform-based Feature Extraction Classifier with Neural Network

Abstract

Fault detection in transmission lines is critical for maintaining the reliability and stability of power systems. This study presents an advanced method for fault detection in transmission line segments using a hybrid approach combining wavelet transform, multiclass Support Vector Machine (SVM), and neural networks. The wavelet transform is employed for feature extraction, allowing for accurate localization and characterization of faults by capturing high-frequency transient signals associated with fault events. These features are then fed into a multiclass SVM classifier to distinguish between different types of faults accurately. To further enhance the classification performance, a neural network is integrated, optimizing the fault detection process by improving the decision-making capabilities of the SVM classifier. Extensive simulation results demonstrate the effectiveness of the proposed approach, showing high accuracy in fault classification under various fault conditions and environmental noise levels. This method provides a robust and reliable tool for real-time monitoring and fault detection in transmission systems, ensuring prompt response and improved system resilience.