在电力现货市场下如何最大化售电商参与需求侧响应的长期收益？

图13 20次需求响应时获得的累积收益Fig. 13 Cumulative gains from 20 demand responses

4 结论

在电力现货市场环境下,为了减少售电商在用电高峰期实时电价过高带来的损失,本文提出了采用基于神经网络的强化学习的售电商动态优化需求响应方案,通过向用户发布需求响应补贴价格来减少用电负荷,以实现其长期收益最大化。考虑了售电商发布的历史补贴价格对用户舒适成本感知的影响,构建了售电商与用户前后时间状态耦合的动态优化问题。通过选取5个工业用户的实际用电数据进行训练,结果表明使用基于神经网络的强化学习算法能够有效完成训练,计算得出的补贴价格可以有效提高售电商的长期收益。

参考文献

[1]杨旭英,周明,李庚银.智能电网下需求响应机理分析与建模综述[J].电网技术,2016,40(1):220-226.YangXuying,ZhouMing,LiGengyin.Survey on demand response mechanism and modeling in smart grid[J].Power System Technology,2016,40(1):220-226(in Chinese).

[2]田世明,王蓓蓓,张晶.智能电网条件下的需求响应关键技术[J].中国电机工程学报,2014,34(22):3576-3589.TianShiming,WangBeibei,ZhangJing.Key technologies for demand response in smart grid[J].Proceedings of the CSEE,2014,34(22):3576-3589(in Chinese).

[3]沈运帷,李扬,高赐威,等.需求响应在电力辅助服务市场中的应用[J].电力系统自动化,2017,41(22):157-167.ShenYunwei,LiYang,GaoCiwei,et al.Application of demand response in ancillary service market[J].Automation of Electric Power Systems,2017,41(22):157-167(in Chinese).

[4]徐筝,孙宏斌,郭庆来.综合需求响应研究综述及展望[J].中国电机工程学报,2018,38(24):84-95+336.XuZheng,SunHongbin,GuoQinglai.Reviewandprospect of integrated demand response[J].Proceedings of the CSEE,2018,38(24):84-95+336(in Chinese).

[5]王剑晓,钟海旺,夏清,等.基于成本-效益分析的温控负荷需求响应模型与方法[J].电力系统自动化,2016,40(5):45-53.WangJianxiao,ZhongHaiwang,XiaQing,et al.Model and method of demand response for thermostatically-controlled loads based on cost-benefit analysis[J].Automation of Electric Power Systems,2016,40(5):45-53(in Chinese).

[6]卫文婷,王丹,贾宏杰,等.一种基于模型预测的城市园区分层分布式温控负荷需求响应控制策略[J].中国电机工程学报,2016,36(8):2049-2056.WeiWenting,WangDan,JiaHongjie,et al.A hierarchical and distributed control strategy of thermostatically controlled appliances for city park based on load model prediction[J].Proceedings of the CSEE,2016,36(8):2049-2056(in Chinese).

[7]陈雨果,张轩,罗钢,等.用户报量不报价模式下电力现货市场需求响应机制与方法[J].电力系统自动化,2019,43(9):179-186.ChenYuguo,ZhangXuan,LuoGang,et al.Demand response mechanism and approach of electricity spot market in bidding mode without price on the user side[J].Automation of Electric Power Systems,2019,43(9):179-186(in Chinese).

[8]艾欣,赵阅群,周树鹏.适应清洁能源消纳的配电网直接负荷控制模型与仿真[J].中国电机工程学报,2014,34(25):4234-4243.AiXin,ZhaoYuequn,ZhouShupeng.Direct load control model and simulation for clean energy accommodation in distribution network[J].Proceedings of the CSEE,2014,34(25):4234-4243(in Chinese).

[9]KaiMa,TingYao,JieYang,et al.Residential power scheduling for demand response in smart grid[J].International Journal of Electrical Power & Energy Systems,2016,78:320-325.

[10]罗琴,宋依群.售电市场环境下计及可中断负荷的营销策略[J].电力系统自动化,2015,39(17):134-139.LuoQin,SongYiqun,Marketing strategy in competitive retail market considering interruptible load[J].Automation of Electric Power Systems,2015,39(17):134-139(in Chinese).

[11]周保荣,黄廷城,张勇军.计及激励型需求响应的微电网可靠性分析[J].电力系统自动化,2017,41(13):70-78.ZhouBaorong,HuangTingcheng,ZhangYongjun.Reliability analysis on microgrid considering incentive demand response[J].Automation of Electric Power Systems,2017,41(13):70-78(in Chinese).

[12]曹佳,马洪艳,刘扬,等.基于节点电价的需求响应策略研究[J].电网技术,2016,40(5):1536-1542.CaoJia,MaHongyan,LiuYang,et al.Research on demand response strategy based on nodal price[J].Power System Technology,2016,40(5):1536-1542(in Chinese).

[13]MaharjanS,ZhuQ,ZhangY,et al.Demand response management in the smart grid in a large population regime[J].IEEE Transactions on Smart Grid,2016,7(1):189-199.

[14]LiuZ,WiermanA,ChenY,et al.Data center demand response: Avoiding the coincident peak via workload shifting and local generation[J].Performance Evaluation,2013,70(10):770-791.

[15]徐业琰,廖清芬,刘涤尘,等..XuYeyan,LiaoQingfen,LiuDichen,et al.(in Chinese).

[16]ChaiB,ChenJ,YangZ,et al.Demand response management with multiple utility companies:a two-level game approach[J].IEEE Transactions on Smart Grid,2014,5(2):722-731.

[17]Qian LP,Zhang YJ,HuangJ,et al.Demand response management via real-time electricity price control in smart grids[J].IEEE Journal on Selected Areas in Communications,2013,31(7):1268-1280.

[18]孙宇军,王岩,王蓓蓓,等.考虑需求响应不确定性的多时间尺度源荷互动决策方法[J].电力系统自动化,2018,42(2):106-113.SunYujun,WangYan,WangBeibei,et al.Multi-time scale decision method for source-load interaction considering demand response uncertainty[J].Automation of Electric Power Systems,2018,42(2):106-113(in Chinese).

[19]张禹森,孔祥玉,孙博伟,等.基于电力需求响应的多时间尺度家庭能量管理优化策略[J].电网技术,2018,42(6):1811-1819.ZhangYusen,KongXiangyu,SunBowei,et al.Multi-time scale home energy management strategy based on electricity demand response[J].Power System Technology,2018,42(6):1811-1819(in Chinese).

[20]张忠会,刘故帅,谢义苗.基于博弈论的电力系统供给侧多方交易决策[J].电网技术,2017,41(6):1779-1785.ZhangZhonghui,LiuGushuai,XieYimiao.A game theory approach to analyzing multi-party electricity trading on supply side[J].Power System Technology,2017,41(6):1779-1785(in Chinese).

[21]刘一欣,郭力,王成山.多微电网参与下的配电侧电力市场竞价博弈方法[J].电网技术,2017,41(8):2469-2476.LiuYixin,GuoLi,WangChengshan.Optimal bidding strategy for microgrids in electricity distribution market[J].Power System Technology,2017,41(8):2469-2476(in Chinese).

[22]尹龙,刘继春,高红均,等.考虑多种用户价格机制下的综合型能源售电公司购电竞价策略[J].电网技术,2018,42(1):88-97.YinLong,LiuJichun,GaoHongjun,et al.Study on bidding strategy of comprehensive power retailer under multiple user-price mechanisms[J].Power System Technology,2018,42(1):88-97(in Chinese).

[23]FahriogluM,Alvarado FL.Using utility information to calibrate customer demand management behavior models[J].IEEE Transactions On Power Systems,2001,16(2):317-322.

[24]朱兆霞,邹斌.PJM日前市场电价的统计分析[J].电力系统自动化,2006,30(23):53-57.ZhuZhaoxia,ZouBin.Statistical analysis of day-ahead prices in PJM market[J].Automation of Electric Power Systems,2006,30(23):53-57(in Chinese).