集成差异化建筑集群的微网系统分层分布式优化调度  

Hierarchical and Distributed Optimal Scheduling for Microgrid System Integrating Differentiated Building Clusters

作　　者：贾宏杰[1] 汪晗忱 靳小龙[1] 穆云飞[1] 陈盛 张家郡 JIA Hongjie;WANG Hanchen;JIN Xiaolong;MU Yunfei;CHEN Sheng;ZHANG Jiajun(Key Laboratory of Smart Grid of Ministry of Education(Tianjin University),Tianjin 300072,China;China Electric Power Research Institute,Beijing 100192,China;Electric Power Research Institute of State Grid Jilin Electric Power Co.,Ltd.,Changchun 130021,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300072 [2]中国电力科学研究院有限公司,北京市100192 [3]国网吉林省电力有限公司电力科学研究院,吉林省长春市130021

出　　处：《电力系统自动化》2024年第22期96-107,共12页Automation of Electric Power Systems

基　　金：国家电网公司科技项目(5700-202229196A-1-1-ZN)。

摘　　要：针对差异化绝热性能建筑集群与微网系统间的协同互动,提出一种基于交替方向乘子法(ADMM)的分层分布式优化调度策略,在实现微网高效、经济运行的同时,充分考虑不同建筑集群的差异化利益诉求和隐私保护需求。首先,考虑多主体利益诉求,构建了集成差异化绝热性能建筑集群的微网系统分层调度模型。在上层,微网系统运营商(MSO)通过优化购电功率、售电功率和售电价格,最大化其收益;在下层,采用热阻-热容网络对建筑的热动态过程进行差异化建模,建筑集群用户通过优化暖通空调系统送风温度和购电价格,最小化其用能成本。然后,考虑数据隐私性需求,采用ADMM进行分层分布式求解。通过引入虚拟变量拆分上层MSO和下层建筑集群用户子问题,并耦合上下层优化变量,根据ADMM进行局部更新和交替迭代,得到全局优化解。为确保算法的收敛性,利用McCormick包络松弛方法将子问题转化为凸优化。最后,通过算例分析验证了分层分布式优化调度策略的合理性和有效性。A hierarchical and distributed optimal scheduling strategy based on the alternating direction method of multipliers(ADMM)is proposed for the collaborative interaction between building clusters with differentiated adiabatic performance and microgrid systems,which fully considers the differentiated interest demands and privacy protection needs of different building clusters while realizing the efficient and economic operation of microgrid.Firstly,a hierarchical scheduling model for the microgrid system integrating building clusters with differentiated adiabatic performance is constructed considering the multi-agent interests.At the upper level,the microgrid system operator(MSO)maximizes its revenue by optimizing the purchased power,sold power,and electricity prices.At the lower level,the thermal dynamic process of buildings is differentially modeled using a resistancecapacitance network,and building cluster users minimize their energy costs by optimizing the supply air temperature of heating,ventilation,and air conditioning(HVAC)systems and electricity prices.Then,considering the requirement for data privacy,the ADMM is employed for hierarchical and distributed solution.By introducing virtual variables to decompose the upper-level MSO and lower-level building cluster user subproblems,and coupling the optimization variables of upper and lower levels,local updates and alternating iterations are conducted based on ADMM to obtain the global optimal solutions.To ensure the algorithm convergence,the McCormick envelope relaxation method is employed to convert subproblems into convex optimization.Case studies are conducted to validate the rationality and effectiveness of the hierarchical and distributed optimal scheduling strategy.

关 键 词：分层分布式优化 智能建筑 微网 交替方向乘子法 建筑热惯性 需求响应 

分 类 号：TM73[电气工程—电力系统及自动化]