REVIEW OF DATA-DRIVEN AND MODEL-BASED PIPELINE MONITORING AND LEAKAGE DETECTION TECHNIQUES

Abstract

Pipeline leakage detection and monitoring systems are crucial for ensuring the safety, efficiency, and reliability of pipeline infrastructure, which is vital for economic growth, environmental protection and public safety. This review provides a comprehensive overview of data-driven and model-based approaches for pipeline leakage detection and monitoring. Existing literatures on advanced data analytics techniques, including machine learning, statistical process control, and model-based methods, such as pressure transient analysis and inverse transient analysis are examine. Furthermore, the review highlights the strengths and limitations of each approach, discusses the challenges associated with pipeline leakage detection, and identifies future research directions and conclude by providing insights that can be adopted for the development of more effective and efficient pipeline leakage detection and monitoring systems, ultimately contributing to the reduction of pipeline failures and environmental impacts.