原文信息:
Strategic retail pricing and demand bidding ofretailers in electricity market: A data-driven chance-constrained programming
原文链接:
https://www.sciencedirect.com/science/article/pii/S266679242200018X
Highlights
•提出了一种混合深度学习的预测方法;
•提出了一个针对零售商、消费者和市场清算过程的双层次模型;
•提出了一个考虑到不确定性的机会约束程序;
•研究了需求灵活性对零售商策略的影响。
Research gap
本文利用了双层优化模型解决了电力零售商在零售市场,日前批发市场,日内平衡市场之间的策略定价和竞拍机制。另外在数据驱动的混合深度学习预测和机会约束程序,电力零售商的用户需求灵活性也得到了研究。
Abstract
This paper proposes a novelbi-level optimization model to study the strategic retail pricing and demandbidding problems of an electricity retailer that considers the interactionsbetween demand response and market clearing process. In order to accuratelyforecast the day-ahead demand bids submitted by the retailer, a novel deeplearning framework based on convolutional neural networks and long short-termmemory is proposed that can capture both local trends and long-term dependencyof the forecasting data. In addition, uncertainties about the retailer’s serveddemand, rivals’ demand bids, and wind power generation are incorporated usingthe data-driven uncertainty set constructed from data. We further proposechance-constrained programming that introduces a set of chance constraints torepresent the operational risk associated with the market uncertainties. Tosolve this problem, we first reformulate chance-constrained programming as atractable second-order conic programming and then convert it into asingle-level mathematical program with equilibrium constraints by using itsKarush Kuhn Tucker conditions. The scope of the examined case studies isfour-fold. First, they evaluate the benefits of the proposed forecastingframework in terms of higher accuracy and expected profit compared to theconventional forecasting methods. Second, they demonstrate how demandflexibility affects the retailer’s strategies and its business cases. Third,they highlight the added value of the proposed bi-level model capturing themarket clearing process by comparing its outcomes against the state-of-the-artbi-level model with exogenous market prices. Finally, they analyze theretailer’s strategies and business cases at different confidence levelsregarding the imposed chance constraints.
Keywords
Electricity retailer 电力零售商
Demand response需求侧响应
Deep learning 深度学习
Bi-level optimization problem 双层优化问题
Chance-constrained programming 机会约束程序
图1 双层优化模型
图2 算法示意图
图3 混合深度学习预测模型
图4 预测结果的比较
图5 零售电价,需求侧响应,竞拍策略,系统负荷,批发市场价格,平衡策略,平衡价格
团队介绍
本研究由英国帝国理工学院、香港大学、以及清华大学的研究人员共同完成。
通信作者简介:
仇大玮,博士,英国帝国理工学院博士后研究员,从事电力市场定价,综合能源系统,和人工智能研究。在Applied Energy、Advances in Applied Energy等期刊上发表论文超过20篇,并多次受邀报告。
董子航,博士,英国帝国理工学院博士后研究员。2019年获得帝国理工学院控制系统博士学位。博士研究课题为鲁棒经济模型预测控制的稳定性理论和性能分析。现主要从事综合能源系统的集中和分布式优化、运行和分析,以及锂电池电化学模型研究。
阮广春,博士,香港大学电力电子工程系博士后研究员,主要从事电力市场、需求响应、机器学习、电力系统运行优化研究。2021年获清华大学博士学位,2019和2020年分别前往美国华盛顿大学和德州农工大学交流访问。
关于Applied Energy
本期小编:柴汝宽;审核人:仇大玮
《Applied Energy》是世界能源领域著名学术期刊,在全球出版巨头爱思唯尔 (Elsevier) 旗下,1975年创刊,影响因子10.109,CiteScore 20,高被引论文ESI全球工程期刊排名第4,谷歌学术全球学术期刊第53,本刊旨在为清洁能源转换技术、能源过程和系统优化、能源效率、智慧能源、环境污染物及温室气体减排、能源与其他学科交叉融合、以及能源可持续发展等领域提供交流分享和合作的平台。开源(Open Access)姊妹新刊《Advances in Applied Energy》现已正式上线。在《Applied Energy》的成功经验基础上,致力于发表应用能源领域顶尖科研成果,并为广大科研人员提供一个快速权威的学术交流和发表平台,欢迎关注!
公众号团队小编招募长期开放,欢迎发送自我简介(含教育背景、研究方向等内容)至wechat@applied-energy.org
点击“阅读原文”
喜欢我们的内容?
点个“赞”或者“再看”支持下吧!