极端环境干扰下电网弹性分析及优化策略研究

Power grid resilience, used to measure the ability of the power grid to resist and recover quickly from the disruptions of natural disasters and manmade attacks, is a critical characteristic of the modern power system. This project aims to study the power grid resilience and its optimization strategies from the perspective of network science. First, the concepts of state transitions will be applied to describe and model the power grid operation process under the extreme conditions. In the model, we will clarify the mechanisms of grid state transitions, as well as consider the actual effects of the external disruptions, the internal electrical parameters and the structure control on grid state transitions. Then, numerical simulations will be adopted to assess the properties of the power grid cases and mathematical statistics methods will be used to analyze these simulation results, to explore the intrinsic relations between the resilience of the power grid and its structure that includes the overall network topological characteristics and the locations of distributed generators. Finally, strategies for network structure adjustment before the extreme perturbations and for network structure control during the resilience process will be designed to improve the resilience of the power grid. This project will help reveal the grid operation mechanisms the during the resilience process and provide theoretical principles and methodological foundations for designing grid resilience optimization strategies.

电网弹性被用来衡量电网在极端天气或人为攻击条件下抵御干扰并及时恢复供电的能力，是现代电力系统的重要性能之一。本项目旨在从网络科学的视角对电力系统弹性及其优化策略展开研究。首先，本项目拟应用状态转换的概念，对在极端环境干扰下电网的实时运行过程进行描述和建模。在该模型中，我们将阐明弹性过程中网络状态转换的机理，并充分考虑外界干扰、系统内部电气量参数和结构控制等因素对系统状态转换的作用。然后，应用对电网算例进行数值仿真评估并对评估结果进行关联性统计分析的方法，研究网络整体的拓扑特征以及分布式发电装置的位置分布等拓扑结构因素与系统弹性的关联。最后，设计干扰来临之前的网络结构调整方案和弹性过程中的系统结构控制策略，来优化电网弹性。本项目有助于揭示弹性过程中电力系统运行的机理，并为优化实际电力系统的弹性性能提供理论依据和方法指导。

电网是当代社会的重要生命线。电网中设备地理上广泛分布且多暴露在野外，极易受到极端事件的影响。本项目旨在从网络科学的视角研究电网的弹性建模、评估及其优化策略。首先，建立了电网弹性模型与评估指标体系：引入复杂自适应和多时间尺度观点，将电网弹性全过程分解为多个阶段；明确了各阶段的主导影响因素，数学描述模型，并提出连锁故障后供电损失量、快速恢复措施后的最大供电能力、整个弹性过程的供电量损失等指标定量评估电网弹性性能；此成果将“电力系统弹性”这一大而复杂的性能分解成为多个小而具体的问题，从而使得后续研究更加有的放矢且更有实际意义。其次，清晰揭示了电网的网络结构特征与其弹性性能的关联，并提出了电网关键元件识别方法：发现了灾后电网存在布雷斯悖论现象，从而揭示电网网络连通性与最大供电能力非正相关；提出了基于贪婪算法和多目标优化的电网关键元件识别方法，揭示了电力元件的弹性重要性与连通性重要性弱相关。最后，提出了电网弹性性能优化措施：针对最大供电能力这一弹性指标，提出了将遗传算法和内点法相结合的双环优化算法；针对整个弹性过程的供电量损失这一弹性指标，提出了结合强化学习和内点法的综合供电恢复策略。本项目研究成果为电网公司建设弹性电网提供了有价值参考。.在此项目资助下，共发表标注本项目资助的SCI期刊论文5篇，EI会议论文3篇，受理专利1项。共培养研究生5名。负责人张曦多次参加或组织相关学术会议和活动，取得了一定的国际国内学术影响力和认可度，包括：在IEEE Circuits and Systems Society旗舰会议IEEE ISCAS 2022 申请到题为 “Intelligent analysis and decision-making of large-scale nonlinear complex systems”的Focus Session，此Focus Session共录取文章8篇，其中6篇被组委会选中推荐到IEEE Transactions on Circuits and Systems II；担任了 IEEE Transactions on Circuits and Systems II 的副编辑；受聘成为中国电机工程学会电力系统专委会学术工作组成员；担任第十七届全国复杂网络会议分论坛主持人。