AI ENABLED FAULT DETECTION, CLASSIFICATION, AND LOCALIZATION IN HIGH VOLTAGE TRANSMISSION LINES USING HYBRID MACHINE LEARNING MODELS

High-voltage transmission lines are critical components of modern power systems, responsible for transmitting electrical energy over long distances. However, these lines are highly vulnerable to faults caused by lightning strikes, insulation failures, equipment malfunctions, and environmental disturbances. Conventional protection methods, such as relay-based and impedance-based techniques, often struggle to provide accurate and fast fault identification under complex and dynamic operating conditions. This paper proposes an advanced hybrid machine learning framework for real-time fault detection, classification, and localization in high-voltage transmission lines. The system utilizes voltage and current signals, which are preprocessed and analyzed using wavelet transform techniques to extract both time-domain and frequency-domain features. A Convolutional Neural Network (CNN) is employed for deep feature learning, while a Random Forest (RF) classifier performs accurate fault classification. Additionally, a regression model is used to estimate the precise fault location. The system is further integrated with an IoT-based hardware implementation using an ESP32 microcontroller for real-time monitoring and fault management. Simulation and experimental results demonstrate high accuracy, robustness against noise, reduced response time, and improved reliability. The proposed approach enhances the resilience and efficiency of smart grid systems.