A Study on the Estimation of Tension of Overhead Wire Using Acceleration Sensor Using Gradient Boosting Regression

This study proposes a novel method for estimating tension in overhead wires of distribution systems using acceleration sensors, crucial for ensuring reliable power supply and proactive asset management. Traditional tension sensors, which are costly and complex to install, are replaced by a more cost-effective and robust alternative. Pearson correlation analysis revealed significant correlations between acceleration sensor data and actual wire tension, enabling the use of Gradient Boosting Regression(GBR) for effective tension estimation. The model achieved a high R2 score of 0.95, demonstrating the potential of acceleration sensor-based systems in maintaining system integrity and enabling real-time monitoring for preemptive maintenance. This method promises enhanced operational efficiency, system longevity, and cost savings. Future research should integrate these findings into comprehensive monitoring systems for improved decision-making.