【摘要】 针对传统基于场景抽样的配电网韧性评估方法计算效率低的问题，提出了一种基于概率图模型（probabilistic graphical model,PGM）的配电网节点存活期望计算方法，旨在提高对配电网在极端灾害场景下韧性评估的计算效率。首先，提出了基于系统节点存活期望的韧性评估方法，分析了该方法的有效性。然后，通过构建配电网的概率图模型，建立节点存活期望方程组，并采用迭代方程的方法求解。最后，综合考虑灾后维修过程，以更完整地模拟配电网受灾及恢复过程，提高中长期灾害下对配电网韧性评估的准确性。算例分析表明，基于概率图的方法在保证计算结果与传统场景抽样方法一致性的同时，显著提高了计算效率，尤其在多时段测试中，该方法能够有效模拟配电网在灾害后的恢复过程，为配电网的中长期灾害风险评估提供了新的工具。更多还原

【Abstract】 To address the low computational efficiency of traditional scenario sampling-based methods for resilience assessment in power distribution networks, this paper proposes a method for deriving the survival expectation of nodes in a power distribution system based on a probabilistic graphical model(PGM). The goal is to enhance the computational efficiency of resilience assessment under extreme disaster scenarios. Firstly, the paper presents a resilience assessment method based on the survival expectation of system nodes and analyzes its effectiveness. Then, by constructing a probabilistic graphical model of the power distribution network,a set of survival expectation equations for the nodes is established and solved using iterative equations. Finally, by comprehensively considering the post-disaster repair process, the method aims to more accurately simulate the impact and recovery process of the power distribution system, thereby improving the accuracy of resilience assessment under medium-and long-term disaster conditions. Case studies demonstrate that while ensuring consistency with the results of traditional scenario-based sampling methods,the probabilistic graphical method significantly improves computational efficiency. Especially in multi-period tests, this method effectively simulates the recovery process of the power distribution network after a disaster, providing a new tool for medium-and long-term disaster risk assessment of power distribution networks.更多还原