考虑韧性提升的输电网灾前预防检修多目标多阶段优化

极端气象灾害易使输电网发生大面积故障。输电网设备的健康程度影响着输电网抵御极端气象灾害物理层面的能力。为进一步提升输电网抗灾能力，减轻经济损失，从输电网韧性提升的角度出发，对输电网灾前预防检修策略进行研究。首先提出将灾害风险抵御力作为输电网灾时韧性评价指标，同时针对极端灾害的时空特性，采用场景分析法来计算韧性指标。根据输电网灾前预防检修的韧性提升、检修风险、经济性的要求，提出输电网灾前预防检修多目标多阶段优化模型，并采用NSGA-II算法与数学规划求解器结合的方法求解。采用IEEE RTS-79算例对优化模型的有效性进行了仿真验证，算例结果表明，文章所提优化模型能得出明显的Pareto前沿以及最优折衷解，以此得到最优灾前预防检修计划，在提升灾时韧性的同时控制灾前预防检修的风险与费用，从而帮助电网运行人员做出科学决策，具有较好的应用价值。

Multi objective and multi-stage optimization of pre disaster preventive maintenance of transmission network considering resilience improvement

Extreme weather disasters are easy to cause large area fault of transmission network. The health degree of transmission network equipment affects the ability of transmission network to resist extreme meteorological disasters. In order to further enhance the transmission network disaster resistance ability and reduce economic losses, this paper studies the pre disaster preventive maintenance strategy of transmission network from the connotation of transmission network resilience improvement. In this paper, firstly, disaster risk resilience is proposed as the evaluation index of transmission network disaster resilience, and the scenario analysis method is used to calculate the resilience index according to the spatial and temporal characteristics of extreme disasters. According to the requirements of toughness improvement, maintenance risk and economy of transmission network pre disaster preventive maintenance, a multi-objective and multi-stage optimization model of transmission network pre disaster preventive maintenance is proposed, and the solution is solved by combining NSGA-II algorithm and mathematical programming solver. The example results show that the proposed optimization model can obtain obvious Pareto frontier and optimal compromise solution, so as to obtain the optimal pre-disaster preventive maintenance plan, which can improve the disaster resilience and control the risk and cost of pre-disaster preventive maintenance, so as to help the power grid operators to make scientific decisions and has good application value.