考虑中长期交易的电力现货市场发电商竞价均衡分析

doi:10.11930/j.issn.1004-9649.202507063

中国电力 ›› 2026, Vol. 59 ›› Issue (2): 127-137.DOI: 10.11930/j.issn.1004-9649.202507063

考虑中长期交易的电力现货市场发电商竞价均衡分析

李晓刚¹(), 刘骐源²(), 冯源昊²(), 吴敏¹, 陈中阳¹, 冯冬涵²()

1. 国家电网有限公司华东分部，上海　200120
2. 上海交通大学电力传输与功率变换控制教育部重点实验室，上海　200240

收稿日期:2025-07-21 修回日期:2025-10-17 发布日期:2026-03-04 出版日期:2026-02-28
作者简介:
李晓刚（1975），男，博士，高级工程师（教授级），从事电力市场与电力交易、电力经济研究，E-mail：13916235981@139.com
刘骐源（2001），男，博士研究生，从事电力市场研究，E-mail：qiyuan_liu@sjtu.edu.cn
冯源昊（1999），男，博士研究生，从事电力市场研究，E-mail：fyh386884223@sjtu.edu.cn
冯冬涵（1981），男，通信作者，博士，教授，从事电力市场研究，E-mail：seed@sjtu.edu.cn
基金资助:
国家电网有限公司华东分部科技项目（52992424000W）。

Equilibrium analysis of generator bidding in the electricity spot market considering medium- and long-term transactions

LI Xiaogang¹(), LIU Qiyuan²(), FENG Yuanhao²(), WU Min¹, CHEN Zhongyang¹, FENG Donghan²()

1. East China Branch of State Grid Corporation of China, Shanghai 200120, China
2. Key Laboratory of Control of Power Transmission and Conversion of the Ministry of Education, Shanghai Jiao Tong University, Shanghai 200240, China

Received:2025-07-21 Revised:2025-10-17 Online:2026-03-04 Published:2026-02-28
Supported by:
This work is supported by Science and Technology Project of East China Branch of SGCC (No.52992424000W).

摘要/Abstract

摘要：

为研究中长期市场交易对现货市场运行的影响，分析发电商在现货市场中的报价策略，提出一种模拟电力现货市场发电商竞价均衡的双层优化模型和多智能体深度强化学习（multi-agent deep reinforcement learning，MADRL）求解算法。引入供需比表征现货市场供需关系，并使用前景理论刻画发电商的有限理性行为特征，以此分析中长期市场交易对现货市场中发电商报价策略的影响。在MADRL求解过程中，将发电商建模为智能体，现货市场出清建模为环境，通过迭代求解得到均衡状态下各发电商的报价策略和现货市场出清价格。以中国东部区域包含8家发电商的实际电力系统为例开展仿真，结果表明，该MADRL算法可以有效求解各发电商的报价策略，准确模拟不同中长期市场设置对现货市场运行的影响。研究结论可为电力交易机构评估发电商竞价行为和制定市场规则提供参考依据。

关键词: 中长期市场, 现货市场, 双层模型, 均衡分析, 强化学习

Abstract:

To examine how medium- and long-term (MLT) market transactions affect spot-market operations and to analyze generators' bidding strategies in the spot market, this paper proposes a bilevel optimization model and a multi-agent deep reinforcement learning (MADRL) algorithm to simulate the bidding equilibrium of generators in the electricity spot market. A supply–demand ratio is introduced to characterize spot-market tightness, and the prospect theory is employed to capture generators' bounded-rational behavior, thereby analyzing the impact of MLT transactions on bidding strategies of generators in the spot market. In the MADRL solution process, generators are modeled as agents and market clearing is modeled as the environment; iterative training yields equilibrium bidding strategies for each generator and the corresponding spot-market clearing prices. A case study on an actual power system in Eastern China involving eight generators demonstrates that the proposed MADRL approach effectively computes generators' bidding strategies and accurately simulates the influence of different MLT market settings on spot-market operations. The findings provide quantitative guidance for power trading institutions to assess strategic bidding and to design coordinated rules for the joint operation of MLT and spot markets.

Key words: medium- and long-term market, spot market, bilevel model, equilibrium analysis, reinforcement learning

李晓刚, 刘骐源, 冯源昊, 吴敏, 陈中阳, 冯冬涵. 考虑中长期交易的电力现货市场发电商竞价均衡分析[J]. 中国电力, 2026, 59(2): 127-137.

LI Xiaogang, LIU Qiyuan, FENG Yuanhao, WU Min, CHEN Zhongyang, FENG Donghan. Equilibrium analysis of generator bidding in the electricity spot market considering medium- and long-term transactions[J]. Electric Power, 2026, 59(2): 127-137.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.202507063

https://www.electricpower.com.cn/CN/Y2026/V59/I2/127

图/表 15

图 1 前景理论-价值函数

Fig.1 Value function of prospect theory

图 2 强化学习基本流程框架

Fig.2 Basic process framework of reinforcement learning

图 3 MADDPG求解电力现货市场均衡框架

Fig.3 Framework of MADDPG for electricity spot market equilibrium

表 1 基于MADDPG算法的双层模型求解流程

Table 1 Bilevel model solution process based on MADDPG algorithm

序号	步骤
1	初始化各智能体的策略网络参数$\theta $和价值网络参数$\omega $
2	初始化各智能体的两个目标网络参数$\theta '$和$\omega '$
3	for ${\text{episode}} = 1$ to $M$ do
4	for $t = 1$ to $T$ do
5	各智能体观测市场状态$s = \left[ {{o_1},{o_2},\cdots ,{o_N}} \right]$
6	各智能体策略网络输出动作$a = {\mu _\theta }\left( s \right)$
7	完成市场出清，获得奖励$r$和下一状态$s'$
8	储存状态转移矩阵$\left( {s,a,r,s'} \right)$至经验回放池
9	if $ {\text{episode}}\text{mod} {f^{\text{c}}} = 0 $ then
10	从经验回放池随机抽取四元组$\left( {s,a,r,s'} \right)$
11	计算目标策略网络${\mu _{\theta '}}$和目标价值网络${Q_{\omega '}}$
12	更新价值网络参数$ \omega \leftarrow \omega - {\alpha _{{\text{critic}}}} {\nabla _\omega }{L_{{\text{critic}}}} $
13	if $ {\text{episode}}\text{mod} {f^{\text{a}}} = 0 $ then
14	更新策略网络参数$ \theta \leftarrow \theta + {\alpha _{{\text{actor}}}} {\nabla _\theta }J $
15	end if
16	end if
17	更新目标价值网络参数$\omega ' = \left( {1 - \tau } \right)\omega + \tau \omega '$
18	更新目标策略网络参数$\theta ' = \left( {1 - \tau } \right)\theta + \tau \theta '$
19	end for
20	end for

图 4 测试系统拓扑

Fig.4 Test system topology

表 2 发电商参数设置

Table 2 Parameter settings of power generators

发电商	所在节点	容量/ MW	$ {a_i} $/ (元·(MW·h)^–2)	$ {b_i} $/ (元·(MW·h)^–1)
G1	1	150	0.175	16.0
G2	2	100	0.200	16.5
G3	3	100	0.160	18.0
G4	8	80	0.220	18.5
G5	9	100	0.180	20.0
G6	11	120	0.175	21.0
G7	18	100	0.225	22.0
G8	19	150	0.200	22.5

图 5 系统日负荷需求曲线

Fig.5 System daily load demand curve

图 6 MADDPG算法各发电商报价策略变化曲线

Fig.6 Variation curves of bidding strategies for each generator based on MADDPG algorithm

图 7 MADDPG算法各发电商利润变化曲线

Fig.7 Variation curves of profits for each power generator based on the MADDPG algorithm

图 8 不同中长期交易电量占比下现货市场价格日曲线

Fig.8 Daily curves of spot market prices under different proportions of electricity volume in medium- and long-term transactions

图 9 不同系统供需比下现货市场价格

Fig.9 Spot market prices under different system supply-demand ratios

图 10 不同中长期交易电量占比下发电商报价及现货价格

Fig.10 Quotes of power generators and spot prices under different proportions of electricity volume in medium and long-term transactions

图 11 不同中长期交易利润下发电商报价和现货市场价格

Fig.11 Quotes of power generators and spot prices under different profits in medium and long-term transactions

图 12 不同风险偏好系数下发电商报价策略变化

Fig.12 Changes of power generators's bidding strategies under different risk preference coefficients

图 13 不同损失厌恶系数下发电商报价策略变化

Fig.13 Changes of power generators's bidding strategies under different loss aversion coefficients

参考文献 30

1	国家发展改革委. 关于加快建设全国统一电力市场体系的指导意见[EB/OL]. (2022-01-18) [2025-3-26]. https://zfxxgk.ndrc.gov.cn/web/iteminfo.jsp?id=18646.
2	国家能源局. 关于推进电力市场建设的实施意见[EB/OL]. (2015-11-30) [2025-3-26]. https://www.nea.gov.cn/2015-11/30/c_134867851.htm.
3	唐成鹏. 引入强化学习的电力市场均衡分析方法及其应用研究[D]. 北京: 华北电力大学, 2022.
	TANG Chengpeng. Equilibrium analysis method and its applications for electricity market using reinforcement learning[D]. Beijing: North China Electric Power University, 2022.
4	刘飞宇, 王吉文, 王正风, 等. 基于两阶段深度强化学习算法的多智能体自由合谋竞价机理研究[J]. 中国电机工程学报, 2024, 44 (12): 4626- 4639.
	LIU Feiyu, WANG Jiwen, WANG Zhengfeng, et al. Study on free joint bidding mechanism in multi-agent environment based on two-stage deep reinforcement learning algorithm[J]. Proceedings of the CSEE, 2024, 44 (12): 4626- 4639.
5	尚静怡, 姜欣, 肖东亮, 等. 边际定价与经济调度深度融合的边际成本比较型竞价模式[J]. 电力系统自动化, 2024, 48 (6): 185- 193.
	SHANG Jingyi, JIANG Xin, XIAO Dongliang, et al. Marginal cost comparison bidding mode with deep fusion of marginal pricing and economic dispatching[J]. Automation of Electric Power Systems, 2024, 48 (6): 185- 193.
6	史俊强, 刘泽宇, 冯冬涵. 一种基于标准化期货合约的中长期电力交易新模式[J]. 中国电力, 2021, 54 (6): 29- 35, 70.
	SHI Junqiang, LIU Zeyu, FENG Donghan. A new mode for medium and long-term electricity trade based on standardized futures[J]. Electric Power, 2021, 54 (6): 29- 35, 70.
7	白云霄, 许遵楠, 季天瑶, 等. 北欧电力衍生品市场对中国电力中长期市场建设的启示[J]. 南方电网技术, 2025, 19 (8): 72- 82.
	BAI Yunxiao, XU Zunnan, JI Tianyao, et al. Inspiration of the Nordic electricity derivatives market for the construction of China's medium and long-term electricity market[J]. Southern Power System Technology, 2025, 19 (8): 72- 82.
8	LI Q R, YANG Z F, YU J, et al. Impacts of previous revenues on bidding strategies in electricity market: a quantitative analysis[J]. Applied Energy, 2023, 345, 121304.
9	KAHNEMAN D, TVERSKY A. Prospect theory: an analysis of decision under risk[J]. Econometrica, 1979, 47 (2): 263- 292.
10	FENG Y H, FENG D H, ZHOU Y, et al. Generation side strategy and user side cost based on equilibrium analysis of the power market under the reliability option[J]. Energy, 2024, 287, 129721.
11	王鹤霏, 蔡国伟, 吴同, 等. 基于供用能双侧多主体博弈互动的综合能源系统优化调度策略[J]. 东北电力大学学报, 2024, 44 (1): 83- 93.
	WANG Hefei, CAI Guowei, WU Tong, et al. Optimization and scheduling strategy for comprehensive energy systems based on bilateral multi-agent game interaction between energy supply and demand[J]. Journal of Northeast Electric Power University, 2024, 44 (1): 83- 93.
12	李亚鹏, 赵麟, 王祥祯, 等. 不确定碳-电耦合市场下梯级水电双层竞价模型[J]. 电力系统自动化, 2023, 47 (20): 83- 94.
	LI Yapeng, ZHAO Lin, WANG Xiangzhen, et al. Bi-level bidding model for cascaded hydropower under uncertain carbon-electricity coupled market[J]. Automation of Electric Power Systems, 2023, 47 (20): 83- 94.
13	段声志, 陈皓勇, 郑晓东, 等. 碳市场背景下发电商竞价策略及电力市场均衡分析[J]. 电测与仪表, 2022, 59 (5): 33- 41.
	DUAN Shengzhi, CHEN Haoyong, ZHENG Xiaodong, et al. Bidding strategy of electricity generation and electricity market equilibrium analysis under the background of carbon market[J]. Electrical Measurement & Instrumentation, 2022, 59 (5): 33- 41.
14	窦春霞, 贾星蓓, 李恒. 基于多智能体的微电网中分布式发电的市场博弈竞标发电[J]. 电网技术, 2016, 40 (2): 579- 586.
	DOU Chunxia, JIA Xingbei, LI Heng. Multi-agent-system-based market bidding strategy for distributed generation in microgrid[J]. Power System Technology, 2016, 40 (2): 579- 586.
15	李鹏, 马溪原, 郭祚刚, 等. 基于双层优化的综合能源服务商博弈策略[J]. 电网技术, 2021, 45 (2): 460- 473.
	LI Peng, MA Xiyuan, GUO Zuogang, et al. Game strategy of integrated energy service providers based on bi-level optimization[J]. Power System Technology, 2021, 45 (2): 460- 473.
16	林晓凡, 曾佳妮, 冯冬涵. 可再生能源消纳责任权重制下电力市场优化决策模型[J]. 电力系统自动化, 2021, 45 (6): 158- 168.
	LIN Xiaofan, ZENG Jiani, FENG Donghan. Optimization decision model of electricity market under renewable portfolio standards[J]. Automation of Electric Power Systems, 2021, 45 (6): 158- 168.
17	魏振, 任杰, 李仪佳, 等. 基于非支配排序遗传算法的虚拟同步光储并网系统动态特性解析与多目标协同控制[J]. 电力科技与环保, 2025, 41 (5): 852- 864.
	WEI Zhen, REN Jie, LI Yijia, et al. Research on three-phase grid-connected system of PV-storage dynamic analysis and multi-objective coordinated control of virtual synchronous photovoltaic storage grid connected system via NSGA[J]. Electric Power Technology and Environmental Protection, 2025, 41 (5): 852- 864.
18	李超英, 檀勤良. 基于智能体建模的新型电力系统下火电企业市场交易策略[J]. 中国电力, 2024, 57 (2): 212- 225.
	LI Chaoying, TAN Qinliang. Market trading strategy for thermal power enterprise in new power system based on agent modeling[J]. Electric Power, 2024, 57 (2): 212- 225.
19	张兴平, 王腾, 张馨月, 等. 基于多智能体深度确定策略梯度算法的火力发电商竞价策略[J]. 中国电力, 2024, 57 (11): 161- 172.
	ZHANG Xingping, WANG Teng, ZHANG Xinyue, et al. Bidding strategy for thermal power generation companies based on multi-agent deep deterministic policy gradient algorithm[J]. Electric Power, 2024, 57 (11): 161- 172.
20	周飞航, 王灏, 王海利, 等. 基于多智能体强化学习的电-碳-绿证耦合市场下多市场主体行为研究[J]. 中国电力, 2025, 58 (4): 44- 55.
	ZHOU Feihang, WANG Hao, WANG Haili, et al. Multi-entity behaviors in electricity-carbon-green certificate coupled markets based on multi-agent reinforcement learning[J]. Electric Power, 2025, 58 (4): 44- 55.
21	LIANG Y C, GUO C L, DING Z H, et al. Agent-based modeling in electricity market using deep deterministic policy gradient algorithm[J]. IEEE Transactions on Power Systems, 2020, 35 (6): 4180- 4192.
22	张继行, 张一, 王旭, 等. 基于多代理强化学习的多新型市场主体虚拟电厂博弈竞价及效益分配策略[J]. 电网技术, 2024, 48 (5): 1980- 1991.
	ZHANG Jihang, ZHANG Yi, WANG Xu, et al. Game bidding and benefit allocation strategies for virtual power plants with multiple new market entities based on multi-agent reinforcement learning[J]. Power System Technology, 2024, 48 (5): 1980- 1991.
23	彭春华, 易泰洵, 孙惠娟, 等. 基于多智能体深度确定性策略梯度的电碳耦合市场发电商均衡竞价策略[J]. 电网技术, 2023, 47 (10): 4229- 4239.
	PENG Chunhua, YI Taixun, SUN Huijuan, et al. Power generator balanced bidding based on multi-agent deep deterministic strategy gradient in electricity-carbon coupling market[J]. Power System Technology, 2023, 47 (10): 4229- 4239.
24	FENG Y H, LIU Q Y, FENG D H, et al. Wind power investment equilibrium guided by reliability option considering electricity-carbon emission trading-trading green certificate joint market[J]. Sustainable Energy, Grids and Networks, 2024, 37, 101275.
25	房欣欣, 杨知方, 余娟, 等. 节点电价的理论剖析与拓展[J]. 中国电机工程学报, 2020, 40 (2): 379- 390.
	FANG Xinxin, YANG Zhifang, YU Juan, et al. Theoretical analysis and extension of locational marginal price[J]. Proceedings of the CSEE, 2020, 40 (2): 379- 390.
26	冯天天, 李晏, 孙晓琪, 等. 大数据驱动下电-碳市场耦合及协同发展研究综述[J]. 智慧电力, 2024, 52 (1): 55- 64.
	FENG Tiantian, LI Yan, SUN Xiaoqi, et al. Review on electricity-carbon market coupling and synergistic development driven by big data[J]. Smart Power, 2024, 52 (1): 55- 64.
27	曹祎, 张莉, 郭静, 等. 基于大语言模型的低碳电力市场发展应用前景[J]. 智慧电力, 2024, 52 (2): 8- 16.
	CAO Yi, ZHANG Li, GUO Jing, et al. Prospects for development of low-carbon electricity markets based on large language models[J]. Smart Power, 2024, 52 (2): 8- 16.
28	何佳蒙, 谢海鹏, 孙啸天, 等. 考虑有限理性的钢铁生产企业电-碳市场交易决策方法[J]. 智慧电力, 2024, 52 (8): 105- 112, 121.
	HE Jiameng, XIE Haipeng, SUN Xiaotian, et al. Bidding strategy for iron-and-steel enterprises in electricity and carbon market considering bounded rationality[J]. Smart Power, 2024, 52 (8): 105- 112, 121.
29	高志远, 曹阳, 邵平, 等. 电力市场领域标准现状和未来发展需求分析[J]. 浙江电力, 2025, 44 (9): 117- 125.
	GAO Zhiyu, CAO Yang, SHAO Ping, et al. Analysis of current status and future standard development requirements in electricity market sector[J]. Zhejiang Electric Power, 2025, 44 (9): 117- 125.
30	何宇斌, 顾慧杰, 彭超逸, 等. 考虑需求响应资源配置的现货-辅助服务市场联合出清机制[J]. 浙江电力, 2025, 44 (9): 126- 136.
	HE Yubin, GU Huijie, PENG Chaoyi, et al. Joint clearing mechanism for the spot and ancillary service markets considering demand response resource allocation[J]. Zhejiang Electric Power, 2025, 44 (9): 126- 136.

[1]	蔺国欣, 张超, 冯凯, 马佳豪, 张宁. 考虑多周期序贯交易风险的年度购电决策方法[J]. 中国电力, 2026, 59(3): 84-93.
[2]	丁一, 王春亮, 席跃伟, 胡文燕, 杨志平, 郭喜燕. 630 MW超临界供汽热电联产机组收益计算[J]. 中国电力, 2026, 59(1): 76-83.
[3]	王奖, 陈晓东, 许喆, 王景亮, 王立鹏. 基于多主体博弈的电力市场不平衡资金分摊优化方法[J]. 中国电力, 2026, 59(1): 153-162.
[4]	吴问足, 王旭辉, 卢苑, 陈婉, 田石金, 陈娴. 基于分形理论的电力现货市场出清电价预测方法[J]. 中国电力, 2025, 58(9): 183-193.
[5]	樊会丛, 段志国, 陈志永, 朱士加, 刘航, 李文霄, 杨阳. 基于多智能体深度策略梯度的离网型微电网双层优化调度[J]. 中国电力, 2025, 58(5): 11-20, 32.
[6]	王力, 蒋宇翔, 曾祥君, 赵斌, 李均昊. 基于深度强化学习的孤岛微电网二次频率控制[J]. 中国电力, 2025, 58(5): 176-188.
[7]	赵彤, 李雪松, 周浩, 丁羽, 杨斌, 王文涛, 王鹏. 基于动态碳排放强度的电碳市场耦合建模方法及市场优化机制分析[J]. 中国电力, 2025, 58(4): 31-43.
[8]	周飞航, 王灏, 王海利, 王萌, 金耀杰, 李重春, 张忠德, 王鹏. 基于多智能体强化学习的电-碳-绿证耦合市场下多市场主体行为研究[J]. 中国电力, 2025, 58(4): 44-55.
[9]	王冠朝, 霍雨翀, 李群, 李强. 基于深度强化学习与改进Jensen模型的风电场功率优化[J]. 中国电力, 2025, 58(4): 78-89.
[10]	李湃, 卢慧, 李驰, 杜洪博. 多能互补发电系统电/热储能容量双层优化配置方法[J]. 中国电力, 2025, 58(3): 55-64.
[11]	曾仪, 周毅, 陆继翔, 周良才, 唐宁恺, 李红. 基于多智能体安全深度强化学习的电压控制[J]. 中国电力, 2025, 58(2): 111-117.
[12]	尚静怡, 张艺涵, 杨福旺, 向明旭. 发电商策略性竞价对省间-省内现货市场影响与应对策略[J]. 中国电力, 2025, 58(11): 1-13.
[13]	宋卓然, 张燕妮, 王阳, 蒋海玮, 李嘉宇, 高洪超. 电力市场环境下分布式能源系统的多元运行模式与灵活性影响分析[J]. 中国电力, 2025, 58(11): 111-121.
[14]	柳华, 熊再豹, 蒋陶宁, 高宇, 金雨含, 葛磊蛟. 分布式强化学习驱动的微电网群动态能量优化管理策略[J]. 中国电力, 2025, 58(10): 50-62.
[15]	李祥光, 谭青博, 李帆琪, 李旭东, 谭忠富. 电碳耦合对煤电机组现货市场结算电价影响分析模型[J]. 中国电力, 2024, 57(5): 113-125.

考虑中长期交易的电力现货市场发电商竞价均衡分析

Equilibrium analysis of generator bidding in the electricity spot market considering medium- and long-term transactions

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 15

参考文献 30

相关文章 15

编辑推荐

Metrics

模态框（Modal）标题

考虑中长期交易的电力现货市场发电商竞价均衡分析

Equilibrium analysis of generator bidding in the electricity spot market considering medium- and long-term transactions

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 15

参考文献 30

相关文章 15

编辑推荐

Metrics