基于 DDPG 算法的可再生能源系统竞标策略-《计算机技术与发展》

文章信息/Info

作者:: 冯伊豪¹; 2 ; 黄超¹; 2 ; 王龙¹; 2 ; 罗熊¹; 2; 1. 北京科技大学计算机与通信工程学院,北京 100083
2. 北京科技大学顺德研究生院,广东佛山 528399

Author(s):: FENG Yi-hao1; 2 ; HUANG Chao1; 2 ; WANG Long1; 2 ; LUO Xiong1; 2; 1. School of Computer and Communication Engineering,University of Science and Technology Beijing,Beijing 100083,China
2. Shunde Graduate School,University of Science and Technology Beijing,Foshan 528399,China

Keywords:: electricity spot market; deep deterministic policy gradient algorithm; renewable energy system; day-ahead bidding; real-timeenergy dispatch

摘要:: 随着中国电力现货市场建设步伐的加快以及大规模可再生能源电厂的涌现,可再生能源参与电力现货市场竞标策略研究对提升电厂的经济效益愈发重要。但可再生能源发电功率与电价的不确定性为其参与电力市场竞标带来了巨大的挑战。为了解决上述问题, 该文提出了基于深度确定性策略梯度 ( DDPG) 算法的风-光-储可再生能源系统日前竞标策略与日内实时能量调度策略。面向日前电力市场,基于日前预测信息,以提升运营收入为目标优化竞标电量,建立日前市场竞标模型;面向实时电力市场,基于超短期预测信息与日前竞标策略,建立日内实时能量调度模型,通过控制蓄电池的充放电过程调整系统的实际供电量。实验结果表明,基于 DDPG 算法的可再生能源系统竞标与能量调度策略可有效提升系统的收入。

Abstract:: Due to the increasing installation of renewable energies and the reformation of electricity market in China,bidding strategy is essential for improving the revenue of renewable energy producers. However, the uncertainty of renewable power generation and electricity price is a great challenge for power bidding. In this study,deep deterministic policy gradient ( DDPG) -based power bidding strategy and energy dispatch strategy are developed for wind-PV-battery integrated energy system in electricity spot market. In day -ahead market,the bidding strategy is developed to optimize the revenue considering day-ahead forecast of renewable power generation.In balancing market,the energy dispatch strategy is developed to control the battery charging / discharging power with updates of forecasts of power generation thereby to adjust the power output from the integrated energy system. Experimental results illustrate that the proposed power bidding and power dispatch strategy can improve the revenue in electricity spot market.