中长期电力市场梯级水电站群多目标对冲模型研究  

Research on Multi-bjective Hedging Model for Cascade Hydropower Stations in Medium-long Term Electricity Market

作　　者：于浩健雄 程春田[1,2] 申建建 唐际政[3] 张俊涛 YU Haojianxiong;CHENG Chuntian;SHEN Jianjian;TANG Jizheng;ZHANG Juntao(Institute of Hydropower&Hydroinformatics,Dalian University of Technology,Dalian 116024,Liaoning Province,China;Key Laboratory of Ocean Energy Utilization and Energy Conservation of Ministry of Education(Dalian University of Technology),Dalian 116024,Liaoning Province,China;PowerChinaZhongnan Engineering Corporation Limited,Changsha 410014,Hunan Province,China)

机构地区：[1]大连理工大学水电与水信息研究所,辽宁省大连市116024 [2]海洋能源利用与节能教育部重点实验室(大连理工大学),辽宁省大连市116024 [3]中国电建集团中南勘测设计研究院有限公司,湖南省长沙市410014

出　　处：《电网技术》2025年第3期1070-1078,I0066,I0067,共11页Power System Technology

基　　金：国家自然科学基金重点项目(52039002);国家博士后创新人才支持计划项目(BX20230055)。

摘　　要：中长期电力市场环境下,如何满足自身市场利益诉求与期末蓄能控制目标成为当下梯级水电站群多目标决策的核心问题之一,其关系到梯级水电站群稳定运行和电网安全。考虑到年内来水不确定性对于梯级水电站群市场利益目标与期末蓄能目标的影响,以及目标间不同量纲的因素,针对性提出2种目标函数:年内来水满足梯级水电站群市场效益与期末蓄能控制需求目标时,以两者优化结果的归一化值最大为目标;反之,以与需求目标差的归一化值平方和最小为目标,利用需求目标之间的“对冲”特性,降低整体损失。首先基于流域历史径流资料,分析来水频率分布特性,通过相关解集法构建符合历史径流特性的场景集作为输入条件;其次根据所提目标函数分别构建非线性优化模型,通过多项式描述梯级水电站发电特性曲线,并实现高效求解;最后根据目标函数条件筛选优化计算结果,得到年可用水量-目标决策对,并分别构建市场效益目标和期末蓄能目标优化决策集。以西南某流域梯级电站群为研究对象,从目标权重系数、与常规调度规则比较、预测电价误差和不同需求目标组合等进行了详细分析,结果表明所提的模型能够在根据来水条件满足需求目标的程度给出最优目标决策;不同权重会影响目标在对冲阶段的侧重;电价误差的不确定性极大地影响对冲阶段的决策,增加了整体的损失;市场目标大小对曲线的对冲阶段范围有较大影响。In the medium-long term electricity market environment,how to meet the demands of its own market benefits and the end of period energy storage control has become one of the core issues of multi-objective decision-making for cascade hydropower stations,which is related to the stable operation of cascade hydropower stations and the safety of the power grid.Considering the impact of the uncertainty of incoming water within the year on the market benefits and end of period energy storage control goals of cascade hydropower stations,as well as the different dimensional factors between the goals,two objective functions are proposed:when the incoming water within the year meets the market benefits and end of period energy storage demand goals of cascade hydropower stations,the goal is to maximize the normalized value of the optimization results of both;On the contrary,with the goal of minimizing the sum of squared normalized values that deviate from the demand objectives,the"hedging"feature between demand objectives is utilized to reduce overall losses.Firstly,based on the historical runoff data of the watershed,analyze the frequency distribution characteristics of incoming water,and construct a scene set that conforms to the historical runoff characteristics as input conditions through the correlation solution set method;Secondly,nonlinear optimization models are constructed based on the proposed objective functions,and the generation characteristic curves of cascade hydropower stations are described through multiple term expressions,achieving efficient solutions;Finally,based on the objective function conditions,the optimization calculation results are selected to obtain the annual available water quantity target decision pair,and the market benefit target and end of period energy storage target optimization decision sets are constructed separately.A detailed analysis was conducted on a cascade power station group in a southwestern watershed,including target weight coefficients,comparison with conventional scheduling

关 键 词：电力市场 最优对冲策略 期末蓄能 梯级调度 非线性规划 

分 类 号：TM721[电气工程—电力系统及自动化]