考虑网架恢复过程的电力应急资源调度方法  

Emergency Power Resource Dispatching Considering Procedure of Power Grid Restoration

作　　者：邓创[1] 刘友波[1] 谭洋洋 余雪莹[1] 刘俊勇[1] 李华强[1] 

机构地区：[1]四川大学电气信息学院,四川省成都市610065 [2]国网成都供电公司,四川省成都市610041

出　　处：《电网技术》2017年第11期3582-3589,共8页Power System Technology

基　　金：国家自然科学基金资助项目(51437003);国家电网公司科技项目(521999150031)~~

摘　　要：大停电后快速恢复系统骨干网架,需制定科学的电力应急资源调度方案。提出一种将节点恢复顺序融入应急资源调度过程的双层决策模型,根据电网恢复原则,上层模型计及了节点恢复顺序和电力应急资源调度方案的影响,并以网架恢复效率最大为目标函数。下层模型按照节点的优先恢复顺序对电力应急资源进行调度,以应急调度成本最小为目标函数。通过应急资源分配变量耦合上下层,引入KKT条件实现双层到单层模型的转换。综合遗传算法和变邻域搜索各自优势,采用混合遗传–变邻域搜索算法求解模型。用新英格兰39节点系统的算例验证了所提模型的合理性。For backbone network restoration after blackout it is necessary to design scientific emergency power resource allocation schedules as fast as possible.In this paper,a bi-level decision model is proposed,integrating node restoration order into emergency resource allocation process.According to restoration criterion,the upper level objective function identifies network restoration efficiency considering effect of node restoration order and emergency power resource allocation schedules.The lower level objective function identifies total cost of emergency allocation and determines emergency power resource allocation schedule based on node priority restoration order.On this basis,by introducing emergency resource allocation variable to couple the upper and lower models,the bi-level model is transformed into a single-level model with KKT condition.Integrating advantages of genetic algorithm（GA） and variable neighborhood search（VNS） and adopting hybrid GA-VNS algorithm,the proposed model is solved.Simulation results on New England 39-bus system validates rationality of the proposed model.

关 键 词：应急资源调度 双层决策 网架恢复 遗传算法 变邻域搜索 

分 类 号：TM72[电气工程—电力系统及自动化]