Abstract: This paper proposes an active distribution network reliability assessment method that takes into account the failure of distribution network automation equipment in multiple scenarios, in view of the "double carbon" target that will promote the development of clean energy access and intelligent distribution network equipment in power systems. Firstly, Latin hypercube sampling method with time series segmentation clustering algorithm is used to generate scenarios. Then, based on the theory of equipment health index to establish a fault model of automation equipment under multiple scenarios, as well as a comprehensive consideration of the source-load model under multiple scenarios, the real-time formation probability of islands and the operation of automation equipment under that time period are calculated, and then the reliability of the distribution network is evaluated and the system reliability index is calculated, and finally the effectiveness of the method and model is proved by the IEEE-RBTS BUS6 F4 arithmetic example.