The distribution network system is directly connected to users, and its faults directly affect people’s daily life and economic activities. In order to realize rapid fault location, isolation and self-healing, this paper carries out a new energy distribution system reliability assessment based on sequential Monte Carlo simulation algorithm. Comparing the use of sequential simulation algorithms, non-sequential simulation algorithms and pseudo-sequential simulation algorithms in the state analysis of distribution network systems, the sequential Monte Carlo method is selected to establish a simulation model for distribution network reliability assessment. According to the “2-3-6” architecture system commonly used in distribution networks, a new energy distribution network self-healing system data communication scheme is designed. The Monte Carlo simulation method is used to model and analyze the reliability of new energy coupled components and to evaluate the reliability indexes. The simulation environment is designed and the RTDS model is used to test the self-healing system of intelligent devices in the distribution network. In the new energy distribution network data scheme designed in this paper, the distribution network self-healing technology is able to remove the fault time within 95ms and complete the fault point elimination work. And when the DG can make the far power side and the switch is located in the near power side, the distribution self-healing system is able to remove the fault point within 60ms time and complete the self-healing operation of the distribution network fault.