城市电网灾后应急资源的集中匹配-分布调度策略  

A Post-Disaster Centralized Matching and Decentralized Dispatch Strategy for Emergency Resources in Urban Power Network

作　　者：万海洋 刘文霞 石庆鑫 刘佳怡 张帅 Wan Haiyang;Liu Wenxia;Shi Qingxin;Liu Jiayi;Zhang Shuai(State Key Laboratory of Alternate Electrical Power System with Renewable Energy Resources North China Electric Power University,Beijing 102206 China)

机构地区：[1]新能源电力系统全国重点实验室(华北电力大学),北京102206

出　　处：《电工技术学报》2024年第23期7463-7480,共18页Transactions of China Electrotechnical Society

基　　金：国家自然科学基金资助项目(52307094)。

摘　　要：城市电网(UPN)是支撑城市功能正常运行的关键基础设施,及时制定UPN灾后应急资源(ER)的最优调度策略对维护城市各生命线系统功能具有重要意义。与单一输、配电网不同,UPN具有多电压等级交杂和拓扑结构复杂等特点,该文针对灾后UPN负荷应急恢复过程中存在ER总量有限、故障元件(FC)数量繁多、多电压等级电网故障元件的协同抢修存在信息壁垒等问题,提出了UPN灾后ER与FC的集中匹配-分布调度策略。首先,根据ER和FC的实时信息和特征参数,采用二分图法建立二者间的匹配关系;其次,以匹配净收益最大为目标,采用改进后的库恩-曼克尔斯算法,集中制定ER和FC间的最优匹配方式;此外,建立计及UPN中多电压等级电网间耦合关系的双层优化模型,其中,上层高压电网以最大化变电站节点负荷恢复量为目标,下层中压电网以最小化负荷损失成本为目标,制定各电压等级负荷的分布式恢复策略;同时,通过及时更新UPN中新探明的FC和ER的信息,重新确定最优匹配方式和负荷恢复策略,保证所提方法在各时间步长内的最优性;最后,算例部分采用实际UPN对该方法进行仿真分析,结果表明该方法能够在有效地降低UPN负荷损失的同时,大幅缩短求解时间。The urban power network(UPN)is a critical infrastructure that supports the normal operation of cities.The timely development of optimal post-disaster emergency resource(ER)allocation strategies for the UPN is of great significance for maintaining the functionality of various lifeline systems in cities.Unlike a single transmission and distribution network,the UPN has the characteristics of multiple voltage levels and complex topologies.This paper proposes a centralized matching and decentralized dispatch strategy for post-disaster ERs in the UPN,aiming at the problem of limited emergency resources,numerous fault components(FC),and information barriers for collaborative repair of FCs among multi-voltage power networks in the post-disaster restoration of the UPN.First,the matching relationship between ER and FC is established using the bipartite graph method based on real-time information and characteristic parameters.Second,with the objective of maximizing the net benefit of matching and using the modified Kuhn-Munkres algorithm,the optimal matching between the ER and the FC is formulated centrally.Additionally,considering the coupling relationship between multi-voltage-level power systems within the UPN,a bi-level optimization model is established.The upper-level high-voltage power system maximizes the load restoration of the substation buses,while the lower-level medium-voltage power system minimizes the load loss cost,and a decentralized recovery strategy for loads of each voltage level is developed.Furthermore,by timely updating the information of newly discovered FC and ER in UPN,the optimal matching method and load restoration strategy are re-determined to ensure that the proposed method is optimal at each time step.Finally,the case study employs real-world UPN data to perform a simulation analysis of the method proposed in this paper.In comparison to commonly used centralized and rolling optimization approaches in existing research,the method proposed in this paper can swiftly formulate solutions that approa

关 键 词：城市电网 极端灾害 故障抢修 负荷恢复 应急资源 赋权二分图 双层优化模型 

分 类 号：TM614[电气工程—电力系统及自动化]