面向大规模异构分布式资源支撑的配电网快速应急恢复多时间尺度滚动优化策略  

Multi-time Scale Rolling Optimization Strategy for Rapid Emergency Recovery of Distribution Network Supported by Large-scale Heterogeneous Distributed Resources

作　　者：刘方 张浩 杨秀 崔勇 徐耀杰 LIU Fang;ZHANG Hao;YANG Xiu;CUI Yong;XU Yaojie(College of Electrical Engineering,Shanghai University of Electric Power,Yangpu District,Shanghai 200090,China;Electric Power Dispatching and Communication Center of State Grid Shanghai Municipal Electric Power Company,Pudong New Area,Shanghai 200122,China)

机构地区：[1]上海电力大学电气工程学院,上海市杨浦区200090 [2]国网上海市电力公司电力调度控制中心,上海市浦东新区200122

出　　处：《电网技术》2023年第8期3311-3323,共13页Power System Technology

基　　金：国家自然科学基金项目(52207121);国家电网有限公司科技项目(5209001200083):“用于输电网关键节点黑启动的分布式资源调控技术研究”;上海市科委青年扬帆计划(21YF1414600)。

摘　　要：在配电网发生大面积停电后利用分布式资源(distributed resource,DER)制定应急恢复方案以实现局部快速自愈,可有效提高其韧性水平。对此,考虑配电网内“源-荷”时变特性对恢复过程的影响,基于模型预测控制框架建立短期全局协调优化和超短期子系统自律调控相配合的快速恢复滚动优化策略。首先,分析配电网停电至恢复过程中面临的多重随机因素,从时间域和空间域解析“源-荷”状态时变特性,形成DER支撑配电网快速恢复的韧性评估指标;然后,基于停电时刻“源-荷”状态预测和子系统划分结果,针对子系统间恢复过程中互为驱动、相互支撑的特征,建立短期快速恢复全局协调优化模型,形成全域范围内的子系统网架结构重构和关键电源、重要负荷恢复方案;针对子系统自治特征,以全局协调优化结果为基准,以恢复过程中负荷停电损失和网损最小为目标,以电源输出功率和负荷投切为控制变量,构建子系统超短期滚动调控策略,将调控结果反馈至全局进行优化改进,形成“源-网-荷-储”最佳能量流匹配的精准调控方案;最后,以改进IEEE33节点配电网系统为例开展仿真分析,结果表明,所提模型和方法显著加快了配电网恢复进程,有效提升其韧性水平。After a large-scale power failure in the distribution network,the distributed resource(DER)is used to formulate an emergency recovery scheme to achieve local rapid self-healing,effectively improving its toughness.In this regard,considering the influence of the"source load"time-varying characteristics in the distribution network on the recovery process,a rolling optimization strategy for rapid recovery is established based on the model predictive control framework.Firstly,the multiple random factors in the process of a distribution network outage and restoration are analyzed,and the time-varying characteristics of"source load"state are analyzed from the time and space domains to form the toughness evaluation index of the DER to support the rapid restoration of the distribution network;Then,based on the"source load"state prediction and the subsystem division results at the time of a power failure,and aiming at the characteristics of mutual driving and support in the recovery process between the subsystems,a short-term rapid recovery global coordination optimization model is established to form the global subsystem grid structure reconstruction and the key power supply and important load recovery scheme;According to the autonomous characteristics of the subsystem,taking the global coordination and optimization results as the benchmark,regarding the minimum load outage loss and network loss in the recovery process as the goal and the power output power and load switching as the control variables,the sub-system ultra-short-term rolling regulation strategy is constructed.The regulation results are fed back to the global for optimization and improvement,forming an accurate regulation scheme of"source network load storage"optimal energy flow matching;Finally,taking the improved ieee33 bus distribution network system as an example,the simulation results show that the proposed model and method significantly accelerate the restoration process of distribution network and effectively improve its resilience level.

关 键 词：配电网 快速恢复 大规模分布式资源 模型预测控制 网架重构 

分 类 号：TM721[电气工程—电力系统及自动化]