摘要
针对发电企业在电力低碳转型过程中,部署可再生能源发电设备的成本问题,研究了一种基于Q-learning的碳-电联合套利策略。利用电力市场和碳市场价格实时波动的特点,在电力市场中在低价时存储电能,高价时卖出电能。在碳市场中,在低价时购入碳排放权。采取Q-learning算法学习碳-电联合套利策略,以欧洲的3个城市为研究对象,仿真结果表明,应用碳-电联合套利策略可提升可再生能源发电售电收入的1%,减少31%购买碳排放权开支,实现最大化套利目标。由于部署可再生能源发电带来的减排效益,使得碳排放开支再次减少10%-20%。通过将碳市场与电力市场相结合套利,使得套利利润得到了显著提升,验证了所提方法的有效性。
A Q-learning-based joint carbon-electricity arbitrage strategy is studied for the cost of deploying renewable energy generation equipment in the power sector during the low-carbon transition of electricity.Taking advantage of the real-time price fluctuations in the electricity and carbon markets,energy storage is used in the electricity market to store electricity at low prices and sell it at high prices.Meanwhile,in the carbon market,carbon credits are purchased when the price is low.The Q-learning algorithm is adopted to learn the joint carbon-electricity arbitrage strategy,and three European cities are used for the study.The results show that the application of the joint carbon-electricity arbitrage strategy can increase the revenue from the sale of renewable energy generation by 1%and reduce the expenditure on the purchase of carbon credits by 31%,which is a good achievement of the goal of maximizing the arbitrage profit.The deployment of renewable power generation can reduce the carbon expenditure by another 10%-20%due to the emission reduction benefits.By combining carbon market arbitrage with power market arbitrage,the arbitrage profit is significantly improved,and the effectiveness of the method is verified.
作者
余运俊
龚海
龚汉城
陈敏
王忠阳
杨林锋
YU Yunjun;GONG Hai;GONG Hancheng;CHEN Min;WANG Zhongyang;YANG Linfeng(a.School of Information Engineering,Nanchang University,Nanchang 330031,China;School of Artificial Intelligence,Nanchang University,Nanchang 330031,China;Jiangxi Zhuoyun Digital Industry Group,Nanchang 330031,China;College of Information Science and Technology,Beijing University of Chemical Technology,Beijing 100010,China)
出处
《实验室研究与探索》
CAS
北大核心
2023年第8期93-98,110,共7页
Research and Exploration In Laboratory
基金
江西省重点研发计划项目(20214BBG74006)
南昌大学教改项目(NCUJGLX-2021-167-93)。
关键词
联合套利
低碳转型
Q学习
电力市场
碳市场
joint arbitrage
low carbon transformation
Q-learning
electricity market
carbon market
