含共享储能的数据中心微网群分布式优化调度  

Distributed Optimal Scheduling of Data Center Microgrids Cluster With Shared Energy Storage

作　　者：王继东[1] 许秋铭 黄婷[2] 孔祥玉[1] WANG Jidong;XU Qiuming;HUANG Ting;KONG Xiangyu(Key Laboratory for Smart Grid of the Ministry of Education(Tianjin University),Nankai District,Tianjin 300072,China;School of Electrical and Electronic Engineering,North China Electric Power University,Changping District,Beijing 102206,China)

机构地区：[1]智能电网教育部重点实验室(天津大学),天津市南开区300072 [2]华北电力大学电气与电子工程学院,北京市昌平区102206

出　　处：《电网技术》2024年第8期3238-3247,I0056,I0058-I0063,共17页Power System Technology

基　　金：国家重点研发计划项目(2022YFB4200703)。

摘　　要：多个数据中心微网可联合成微网群以实现不同数据中心微网的资源联动和能量互补。首先,建立了含共享储能的数据中心微网群运行模型,考虑了不同数据中心微网间的功率交互、数据负载交互、碳配额交互及共享储能容量使用权分配等耦合约束。其次,提出了考虑时变容量使用权和旋转备用服务的共享储能容量分配与运行策略,以进一步提高共享储能的效能。然后,采用高斯混合模型对数据中心微网内可再生发电的预测误差概率分布进行精确拟合,并结合机会约束处理预测误差的不确定性。最后,通过交替方向乘子法实现各个数据中心微网的独立求解,并引入动态乘子更新策略和预测-矫正因子以提高算法的收敛性。所提模型与方法的有效性在仿真中得到验证。Multiple data center microgrids can be integrated into a microgrid cluster to achieve complementary advantages among different data center microgrids.Firstly,the operation model of the data center microgrids cluster is established,and the coupling constraints of the exchanging of power,IT workload,and carbon allowance,as well as the capacity allocation of shared energy storage,are also considered.Secondly,the strategy of shared energy storage,which considers the time-varying capacities demand and the reserve capacities from adjustable charging and discharging power,is proposed to improve the efficiency of shared energy storage.Then,a Gaussian mixture model is used to accurately describe the probability distribution of prediction error for renewable distributed generation of the data center microgrid,and the uncertainties are addressed by chance constraint.Lastly,the alternating direction multiplier method is applied to realize the independent scheduling of each data center microgrid,and an automatic penalty updating strategy,as well as the predictor-corrector factor,are introduced to improve the convergence of the algorithm.The effectiveness of the proposed model and algorithm has been verified in the case study.

关 键 词：数据中心微网 多资源交互 共享储能 分布式优化 高斯混合机会约束 

分 类 号：TM721[电气工程—电力系统及自动化]