不确定环境下含云计算数据中心的电网韧性增强调度 = Resilience-enhanced scheduling of power system with cloud computing data centers under uncertainty

• Main Authors: 赵天阳 Zhao, Tianyang, 张华君 Zhang, Huajun, 徐岩 Xu, Yan, 王鹏 Wang, Peng
• Other Authors: School of Electrical and Electronic Engineering
• Format: Article
• Language: Chinese
• Published: 2021
• Subjects: Engineering::Electrical and electronic engineering; 韧性 Resilience; 云计算数据中心 Cloud Computing Data Center
• Online Access: https://hdl.handle.net/10356/152922
• Institution: Nanyang Technological University

为解决飓风来临前路径不确定时输电线路随机故障等带来的难题，提出了适用于含云计算 数据中心的电网韧性增强日前调度策略，并将其构建为两阶段风险规避的分布鲁棒优化问题。以 飓风对输电线路的时空影响为出发点，采用蒙特卡洛模拟获得飓风路径不确定时线路的离散故障 集合，并构建基于L1 距离度量的分布鲁棒模糊集合。然后，在日前调度中，对机组和数据中心进行优化以平衡经济性和电网韧性，并采用追索问题量化其对日间调度的影响，形成两阶段优化问题。随后，对优化问题进行确定性转换与解耦求解。最后，以含4 个数据中心的IEEE-RTS 系统为测试算例，验证了所提韧性增强策略应对模糊不确定性的有效性。 To manage the possible transmission line failures under uncertain hurricane tracks before its advent, a day-ahead resilience-enhanced scheduling scheme is proposed for power systems with cloud computing data centers. The scheme is formulated as a two-stage risk aversion distributionally robust optimization problem. Considering the spatial and temporal impacts of hurricanes on transmission lines, a discrete line failure set is generated by the Monte-Carlo simulation scheme, in which the hurricane path uncertainty is considered. This set is further formulated as a distributionally robust ambiguity set using L1 norm distance. In the day-ahead scheduling, the generators and data centers are scheduled to balance the operational efficiency and resilience, considering the impacts of day-ahead scheduling on intra-day scheduling, which is formulated as a recourse problem, and resulting in a two-stage optimization problem. It is reformulated to its robust counterpart and solved by decomposition algorithms. Finally, simulations are conducted on a modified IEEE reliability test system with 4 data centers, and the results verify the effectiveness of the proposed resilience-enhanced strategy in addressing the ambiguity uncertainty.