Abstract: Petal distribution network has extremely high power supply reliability, but its multi-loop closed operation characteristics in the face of multiple faults lead to a lack of adaptive algorithms for formulating repair strategies. To address this issue, an optimization model for multi-fault repair objectives in distribution networks that takes into account the operational characteristics of dual petal distribution networks is established. Based on the repair time and load level, a non-dominated sorting genetic algorithm-Ⅱ is proposed for improving the ring network. The intelligent optimization algorithm is applied to solve the repair plan in the dual petal ring network configuration. Finally, through simulation, it is verified that the proposed algorithm performs more efficiently in formulating repair and recovery strategies, and is suitable for practical repair work.