Optimizing Cable Replacement Schedules in Power Grids: A Data‐Driven Approach with Machine Learning and Power Flow Analysis

Ensuring stable power supply demands maintaining power facilities in a reliable operating condition. Power facilities are categorized into those that can be monitored in real time and those that cannot. Facilities that cannot be monitored, such as cables, are often replaced when they are old or replaced in batches. This replacement method overlooks the actual condition and failure impact of these facilities, potentially resulting in economic losses. In this study, we propose a method for optimizing the cable replacement schedules. This method takes into account the Health Index level and the importance in power grid operation. Leveraging machine learning with a dataset of cables, we predicted the Health Index for each cable. Then, we assessed the importance of each cable by calculating load shedding scenarios through power flow analysis, should a cable fail. Finally, the prioritization for cable replacement was determined based on the potential loss caused by cable failure. © 2025 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.