考虑抽水蓄能电站参与容量交易辅助服务的双层优化策略  

Bi-level Optimization Strategy Considering the Participation of Pumped Storage Power Plants in Auxiliary Services of Capacity Trading

作　　者：唐良 蔡晔 唐夏菲 曹一家[1] 谭玉东 TANG Liang;CAI Ye;TANG Xiafei;CAO Yijia;TAN Yudong(State Key Laboratory of Disaster Prevention&Reduction for Power Grid(Changsha University of Science&Technology)Changsha 410004,Hunan Province,China;State Grid Hunan Electric Power Company Limited Economic&Technical Research Institute,Changsha 410004,Hunan Province,China;Hunan Key Laboratory of Energy Internet Supply-demand and Operation,Changsha 410004,Hunan Province,China)

机构地区：[1]电网防灾减灾全国重点实验室(长沙理工大学),湖南省长沙市410004 [2]国网湖南省电力有限公司经济技术研究院,湖南省长沙市410004 [3]能源互联网供需运营湖南省重点实验室,湖南省长沙市410004

出　　处：《电网技术》2025年第4期1469-1478,I0046-I0050,共15页Power System Technology

基　　金：国家自然科学基金项目(52277076,52307081)。

摘　　要：抽水蓄能电站是重要的调峰调频的资源。目前抽水蓄能电站在运行中始终处于被动调度,其成本由两部制电价弥补,导致抽蓄参与辅助服务市场的积极性不高。新能源发展带来了强不确定性与间歇性,亟须加快确立抽水蓄能电站的独立市场主体地位,促进新能源的稳定消纳。设计了一种抽蓄作为独立主体同时参与辅助服务市场及日前市场的运行机制,建立了捆绑容量优化-日前调度双层优化模型,实现抽蓄、风光合作及与电网的友好互动。上层构建风险效用函数量化抽蓄收益期望底线与风光捆绑容量的函数关系,定义抽蓄与风光合作的理论判据,并建立讨价还价博弈模型进行双方收益的合理分配,以风光收益期望最大为目标进行捆绑容量优化;下层考虑多电源类型结构,以社会发电成本最小为目标,模拟日前市场抽蓄出力、工况、日前收益对容量交易的影响。最后,以某省2020年实际数据为例,用粒子群双层算法求解双层优化模型得到最优捆绑容量、收益分配方案,测算结果表明在有效降低风光出力偏差的同时,抽蓄、风光收益均能显著提高,充分发挥了灵活调节资源的调节作用。The pumped storage power station is an important peak and frequency regulation resource.At present,the pumped storage power station is always in passive dispatching during operation,and its benefits are compensated by the two-part electricity price,which leads to the low enthusiasm of pumped storage to participate in the auxiliary service market.The high proportion of new energy has brought strong volatility and intermittence.It is urgent to speed up the establishment of independent market main body status of flexible regulation resources such as pumped storage power stations to promote the stable consumption of new energy.This paper designs the operation mechanism of pumping and storage as an independent subject to participate in the auxiliary service market and the day-ahead market at the same time and establishes a bi-level optimization model of bundled capacity optimization-day-ahead dispatching to realize pumping and storage,wind-solar system cooperation and friendly interaction with the power grid.The upper layer constructs a risk-utility function to quantify the functional relationship between the expected bottom line of the pumped-storage income and the wind-solar bundle capacity,defines the theoretical criterion for the pumped-storage decision to cooperate with the wind-solar,and establishes a bargaining game model to allocate the cooperative income of the two parties reasonably.The bundle capacity is optimized to maximize the expected wind-solar benefit;the lower layer considers the multi-power type structure and aims to minimize the system power generation cost,and simulate the impact of day-ahead market pumping output,operating conditions,and day-ahead revenue on capacity trading.Finally,taking the actual data of a certain province in 2020 as an example,the double-layer optimization model is solved using the particle swarm bi-level algorithm to obtain the optimal bundled capacity and profit distribution scheme.The calculation results show that while effectively reducing the deviation of wind and solar

关 键 词：联合运行 抽水蓄能 容量优化 收益分配 辅助服务 

分 类 号：TM721[电气工程—电力系统及自动化]