Abstract: As the development of new power systems continues to advance,the distribution network architecture has become increasingly complex,rendering traditional fault monitoring and inspection methods inadequate for meeting the core requirements of safe and stable distribution network operation.Addressing this challenge,the study established a comprehensive autonomous drone inspection technology framework tailored to distribution network fault monitoring scenarios.It focused on four key modules:autonomous flight and swarm formation algorithms,multi-source sensor data fusion,intelligent fault hazard identification model training,and inspection equipment optimization.Through engineering trials conducted in typical mountainous distribution network environments,the effectiveness of the technology was quantitatively validated.The research findings demonstrate that this framework effectively addresses the limitations of traditional inspection methods—including low efficiency,difficulty in identifying hazards,and slow fault localization—and provides a viable practical pathway for the intelligent upgrading of distribution network fault monitoring systems.