ARCHIVES
Deep Learning-Based Fault Detection and Condition Monitoring in High-Voltage Electrical Networks
1 2 UG Scholar, Department of Electrical and Electronics Engineering, Holy Mary Institute of Technology and Science, Hyderabad, Telangana, India. 3 Assistant Professor, Department of Electrical and Electronics Engineering, Holy Mary Institute of Technology and Science, Hyderabad, Telangana, India.
Published Online: May-August 2025
Pages: 19-22
High-voltage electrical networks are critical components of modern power systems, ensuring reliable transmission and distribution of electricity across vast geographical regions. Faults occurring in transmission lines, transformers, circuit breakers, and substations can lead to service interruptions, equipment damage, economic losses, and safety hazards. Traditional fault detection and condition monitoring techniques often rely on manual inspections, rule-based systems, and conventional signal processing methods, which may be inadequate for handling the complexity and volume of data generated by modern power networks. Recent advancements in deep learning have introduced intelligent solutions capable of improving fault diagnosis accuracy and enabling real-time condition monitoring. This study presents a deep learning-based framework for fault detection and condition monitoring in high-voltage electrical networks. The proposed system integrates sensor data acquisition, feature extraction, deep neural network models, and predictive analytics to identify abnormal operating conditions and assess equipment health. Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) networks, and hybrid deep learning architectures are employed to analyze electrical parameters such as voltage, current, temperature, vibration, and partial discharge signals. Performance evaluation demonstrates that deep learning models significantly outperform traditional machine learning techniques in fault classification accuracy, response time, and predictive maintenance capabilities. The proposed framework enhances system reliability, reduces maintenance costs, and supports proactive asset management. The findings highlight the potential of deep learning technologies in developing intelligent and resilient power transmission infrastructures for future smart grid environments.
