SMART ELECTRIC POLE FAULT DETECTION USING IOT AND MACHINE LEARNING

Increasing complexity in modern electrical distribution networks demands intelligent, real-time fault monitoring mechanisms to enhance reliability and operational efficiency. Conventional protection systems predominantly rely on threshold-based relays, manual inspection, and delayed reporting procedures, resulting in extended outage duration and elevated maintenance costs. An IoT-enabled smart electric pole monitoring framework integrated with machine learning–based fault classification is presented to address these limitations. The system incorporates distributed sensors to continuously measure voltage, current, temperature, ambient light intensity, and conductor continuity at pole level. An ESP32-based embedded controller performs real-time data acquisition, preprocessing, and secure wireless transmission to a cloud platform for centralized analysis. Extracted electrical features are processed using supervised learning algorithms to accurately classify abnormal conditions including overvoltage, overcurrent, overheating, and line disconnection. Experimental evaluation demonstrates high classification accuracy, reduced false alarm rates, and significant improvement in response time compared with conventional monitoring approaches. Cloud-based visualization and historical data logging further enable predictive maintenance and scalable deployment across urban and rural infrastructures. The proposed architecture strengthens fault resilience, enhances safety, and supports intelligent smart grid transformation through data-driven decision-making.