A review of data-driven leakage diagnosis methods across pipeline and energy transportation system

The growing demand for energy transmission is driving the evolution of pipeline systems from single transmission to multifunctional, coupled energy transportation system (ETS). This increasing complexity, combined with the inherent safety hazards of leakage, underscores the critical need for real-time monitoring and accurate diagnostics of ETS. While data-driven methods have demonstrated significant advantages in leakage diagnosis and attracted extensive research, a systematic review spanning pipeline to complex ETS remains lacking. To address this gap, this paper systematically reviews recent advances in data-driven leakage diagnosis for pipeline and complex ETS. The advantages of data-driven methods for leakage detection are first elaborated, followed by a detailed discussion of detection methodologies under complex scenarios. Subsequently, two leakage localization methods—based on time difference of arrival and attenuation model matching, respectively—are examined to illustrate their suitability. Finally, considering the limitations of current research, this paper outlines prospective directions for the future development of leakage diagnosis methods.