Distributed acoustic sensing (DAS) offers a transformative route to online monitoring of partial discharges (PD) in high-voltage power cables, overcoming intrinsic limitations of conventional methods in distributed coverage and immunity to electromagnetic interference. This review organizes recent advances of φ-OTDR/coherent-OTDR (COTDR)-based DAS into a triadic framework—performance breakthroughs, engineering bottlenecks, and development strategies. We synthesize evidence showing that innovations in source modulation, channel multiplexing, and sensing–coupling have delivered step-change improvements in detection sensitivity, effective monitoring range, and signal response bandwidth, thereby enhancing the identification of incipient PD events and their precise localization. Building on these results, we analyze persisting technical hurdles, including coherent fading, robust demodulation under low SNR, and constraints in field deployment and system integration. Looking ahead, we outline promising directions centered intelligent signal processing and learning-based diagnostics, and edge computing architectures for real-time inference. The goal is to inform the development of a life-cycle, data-driven paradigm for intelligent sensing and predictive maintenance of high-voltage cable assets, providing both a theoretical basis and forward-looking guidance for future engineering practice.

distributed acoustic sensing (DAS),high-voltage cable faults,partial discharge (PD) monitoring