Harnessing Big Data for Effective Tower Impedance Management in Power Systems

Transmission line reliability is critical for modern power systems, with tower footing impedance playing a pivotal role in minimizing failures caused by lightning surges and fault currents. This study introduces a scalable big data framework for predictive maintenance, integrating advanced analytics and machine learning to address challenges such as real-time data processing, fragmented insights, and data retention. Leveraging a hierarchical data model and centralized storage using MongoDB Atlas, the system efficiently processes tower footing impedance measurements. Random Forest algorithm is employed to classify impedance abnormalities, achieving a 99.36% accuracy. The methodology is validated through a pilot project with the a State Transmission Utility, demonstrating significant reductions in failure rates and enhanced operational efficiency. This study provides a robust foundation for proactive transmission line maintenance, offering a pathway to sustainable and resilient power systems.