Artificial intelligence-based pipeline leakage detection in water distribution networks for liquid-cooled battery thermal management systems

Water leakage presents a critical challenge in water distribution systems, leading to substantial water loss, energy wastage, and operational inefficiencies. In the context of Liquid-Cooled Battery Thermal Management Systems (LC-BTMS), maintaining the integrity of water pipelines is essential to ensure consistent cooling performance and battery safety. Early detection of pipeline leaks is crucial to mitigate these losses and enhance the reliability of battery thermal management. However, the lack of comprehensive data on water transportation networks complicates this task. This study leverages Artificial Intelligence (AI) and Machine Learning (ML) techniques to predict and locate leaks in underground pipelines within water distribution networks (WDNs) that are integral to LC-BTMS operations. By utilizing an open-source machine learning platform, municipalities and industries can deploy customized detection models without relying on costly instrumentation. This research compares the performance of several supervised ML algorithms using real-time data collected through Supervisory Control and Data Acquisition (SCADA) systems, offering a practical and cost-effective solution for early leak detection, ensuring system stability and performance in LC-BTMS.