考虑条件风险价值的“新能源+储能”电站现货市场投标策略  

Spot Market Bidding Strategy for a“Renewable Energy+Energy Storage”Power Station Considering Conditional Value at Risk

作　　者：王洪亮 孙伟卿 桑丙玉 WANG Hongliang;SUN Weiqing;SANG Bingyu(School of Mechanical Engineering,University of Shanghai for Science and Technology,Yangpu District,Shanghai 200093,China;Jiangsu Energy Storage Variable Current and Application Engineering Technology Research Center(China Electric Power Research Institute),Nanjing 210003,Jiangsu Province,China)

机构地区：[1]上海理工大学机械工程学院,上海市杨浦区200093 [2]江苏省储能变流及应用工程技术研究中心(中国电力科学研究院有限公司),江苏省南京市210003

出　　处：《电网技术》2023年第9期3620-3628,共9页Power System Technology

基　　金：国家自然科学基金项目(51777126)。

摘　　要：新能源电站的发电和现货市场电价均存在较大的不确定性,使得其投标决策困难,容易因实际发电量与投标电量之间或实际电价与预测电价之间存在较大偏差而产生惩罚费用或影响投标决策有效性。考虑上述因素,将“新能源+储能”电站作为价格接受者,以国内双结算模式和偏差考核为背景制定现货市场最优投标策略。首先,采用最小门控记忆网络–分位数回归(minimal gated memory network-quantile regression,MGM-QR)组合模型模拟电站出力和电价的多重不确定性,以条件风险价值(conditional value at risk,CVaR)为风险度量措施,构建“新能源+储能”电站的日前市场投标和实时市场预期投标的电量组合模型。其次,在实时市场阶段,基于电价和出力的超短期预测结果,利用鲁棒滚动优化模型调度储能出力并重新确定投标电量以获取最大化收益。最后,通过某地区的实际数据进行验证,结果表明所提方法可以有效满足不同风险偏好决策者投标决策需求,保障电站收益。The power generation of renewable energy power stations and the electricity prices in the spot markets have a great uncertainty,which makes the bidding decisions difficult,and it is easy to generate penalty costs or bad effects on the bidding decisions due to the large deviations between the actual power generation and the bidding power or between the actual electricity prices and the predicted electricity prices.Taking the above factors into consideration,the"renewable energy+energy storage"power station is used as the price receiver,and the optimal bidding strategy in the spot market is formulated under the background of the domestic double settlement model and the deviation assessment.First,the minimal gated memory network-quantile regression(MGM-QR)combination model is used to simulate the multiple uncertainties of the power station output and the electricity price.Taking the conditional value at risk(CVaR)as the risk measure to construct the electricity combination model of the day-ahead market bidding and the real-time market expected bidding of"renewable energy+energy storage"power stations.Second,in the real-time market stage,based on the ultra-short-term forecast results of the electricity price and output,a robust rolling optimization model is used to schedule the energy storage output and re-determine the bidding capacity to maximize the revenue.Finally,through the verification of the actual data in a certain area,the results show that the proposed method effectively meets the bidding decision-making needs of decision makers with different risk preferences,and ensures the income of the power station.

关 键 词：“新能源+储能”电站 不确定性 条件风险价值 现货市场 滚动优化 

分 类 号：TM721[电气工程—电力系统及自动化]