Coordinated Trading Decision-making for Aggregators in Wholesale and Retail Electricity Markets Based on Large Language Models

GAO Junlu; WANG Yiheng; ZHU Yiting; HUANG Gang; YANG Qiang

doi:10.7500/AEPS20250909004

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

Coordinated Trading Decision-making for Aggregators in Wholesale and Retail Electricity Markets Based on Large Language Models

更新时间：2026-09-15

- Coordinated Trading Decision-making for Aggregators in Wholesale and Retail Electricity Markets Based on Large Language Models
- Automation of Electric Power Systems Pages: 1-12(2026)
- 作者机构：
  
  1.浙江大学工程师学院，浙江省杭州市 310015
  2.浙江大学电气工程学院，浙江省杭州市 310027
- 作者简介：
- 基金信息：
  
  National Natural Science Foundation of China(52177119)
- DOI：10.7500/AEPS20250909004
  CLC：
- Received：09 September 2025，
  
  Revised：2026-04-23，
  
  Online First：13 July 2026，
- 稿件说明：
移动端阅览
高君璐,王艺衡,祝怡婷等.基于大语言模型的聚合商批发-零售市场协同交易决策[J/OL].电力系统自动化, http://dx.doi.org/10.7500/AEPS20250909004.

GAO Junlu,WANG Yiheng,ZHU Yiting,et al.Coordinated Trading Decision-making for Aggregators in Wholesale and Retail Electricity Markets Based on Large Language Models[J/OL].Automation of Electric Power Systems, http://dx.doi.org/10.7500/AEPS20250909004.
高君璐,王艺衡,祝怡婷等.基于大语言模型的聚合商批发-零售市场协同交易决策[J/OL].电力系统自动化, http://dx.doi.org/10.7500/AEPS20250909004. DOI：

GAO Junlu,WANG Yiheng,ZHU Yiting,et al.Coordinated Trading Decision-making for Aggregators in Wholesale and Retail Electricity Markets Based on Large Language Models[J/OL].Automation of Electric Power Systems, http://dx.doi.org/10.7500/AEPS20250909004. DOI：

摘要

在新型电力系统和电力市场加速发展背景下，资源聚合商已成为参与电力市场的重要主体。聚合商如何协调批发市场购电与零售市场售电环节中的交易决策是其核心问题之一。然而，传统数学规划方法高度依赖精确模型，且易出现维度灾难导致计算复杂度高。同时，数据驱动的机器学习模型存在可解释性差和对训练数据依赖度高的难题，导致现有方法难以有效兼顾决策透明性与环境适应性。为此，文中提出一种基于大语言模型（LLM）的聚合商批发-零售市场协同交易决策方法。该方法通过基于思维链和心智理论建模的推理机制，实现了结构化市场数据与非结构化外部信息的融合利用，提升了交易决策模型在动态市场交易环境下的适用性。仿真算例验证结果表明，所提方法在多场景下较保守型和激进型交易策略在电力市场适应性和聚合商交易收益方面具有优势。

Abstract

In the background of the rapid development of new power systems and electricity markets

resource aggregators are playing an important role as electricity market participants. One of the central challenges of aggregators lies in making coordinated trading decisions across wholesale electricity procurement and retail electricity sales. However

conventional mathematical programming methods depend heavily on accurate models and are prone to the curse of dimensionality

resulting in high computational complexity. Meanwhile

the data-driven machine learning models suffer from poor interpretability and heavy dependence on training data

which renders existing methods unable to balance decision transparency and environmental adaptability effectively. Therefore

this paper proposes a coordinated trading decision-making method for aggregators in the wholesale and retail electricity markets based on large language models (LLMs). The proposed method leverages chain-of-thought (CoT) and theory-of-mind (ToM) modeling based reasoning mechanisms to realize the integrated utilization of structured market data and unstructured external information

thereby improving the applicability of the trading decision-making model in the dynamic market trading environments. Verification results of simulation cases show that the proposed method outperforms both conservative and aggressive trading strategies in terms of electricity market adaptability and aggregator trading revenues in multiple scenarios.

关键词

Keywords

references

张筱慧，潘永超，张璐，等 . 考虑综合能源聚合商与配电公司利益均衡的配电网故障恢复方法［J］. 电力自动化设备， 2023 ， 43 （ 3 ）： 38 - 45 .

Zhang Xiaohui ， Pan Yongchao ， Zhang Lu ， et al . Distribution network fault recovery method considering balance of interests between integrated energy aggregators and power distribution company ［J］. Electric Power Automation Equipment ， 2023 ， 43 （ 3 ）： 38 - 45 .

赵昊天，王彬，潘昭光，等 . 支撑云-群-端协同调度的多能园区虚拟电厂：研发与应用［J］. 电力系统自动化， 2021 ， 45 （ 5 ）： 111 - 121 .

Zhao Haotian ， Wang Bin ， Pan Zhaoguang ， et al . Research and application of park-level multi-energy virtual power plants supporting cloud-cluster-end multi-level synergetic dispatch ［J］. Automation of Electric Power Systems ， 2021 ， 45 （ 5 ）： 111 - 121 .

黄宇翔，黄蔚亮，牛振勇，等 . 虚拟电厂市场机制与盈利模式及欧洲运行实例分析［J］. 电力系统自动化， 2025 ， 49 （ 9 ）： 17 - 30 .

Huang Yuxiang ， Huang Weiliang ， Niu Zhenyong ， et al . Market mechanisms and profit modes of virtual power plants and analysis of their operation cases in Europe ［J］. Automation of Electric Power Systems ， 2025 ， 49 （ 9 ）： 17 - 30 .

王伟韬，王旭，蒋传文，等 . 考虑夏普比率的虚拟电厂日前投标策略［J］. 电网技术， 2023 ， 47 （ 4 ）： 1512 - 1522 .

Wang Weitao ， Wang Xu ， Jiang Chuanwen ， et al . Day ahead bidding strategy for virtual power plants considering Sharpe ratio ［J］. Power System Technology ， 2023 ， 47 （ 4 ）： 1512 - 1522 .

范文轩，袁至，王维庆，等 . 绿氢储能系统综合评价体系和双层优化方法［J］. 电力自动化设备， 2025 ， 45 （ 4 ）： 19 - 27 .

Fan Wenxuan ， Yuan Zhi ， Wang Weiqing ， et al . Comprehensive evaluation system and double-layer optimization method of green hydrogen energy storage system ［J］. Electric Power Automation Equipment ， 2025 ， 45 （ 4 ）： 19 - 27 .

Zugno M ， Morales J M ， Pinson P ， et al . A bilevel model for electricity retailers' participation in a demand response market environment ［J］. Energy Economics ， 2013 ， 36 ： 182 - 197 .

赵冬梅，陶然，马泰屹，等 . 基于多智能体深度确定策略梯度算法的有功-无功协调调度模型［J］. 电工技术学报， 2021 ， 36 （ 9 ）： 1914 - 1925 .

Zhao Dongmei ， Tao Ran ， Ma Taiyi ， et al . Active and reactive power coordinated dispatching based on multi-agent deep deterministic policy gradient algorithm ［J］. Transactions of China Electrotechnical Society ， 2021 ， 36 （ 9 ）： 1914 - 1925 .

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 .

Liu O ， Fu D Q ， Yogatama D ， et al . DeLLMa： decision making under uncertainty with large language models ［PP/OL］. ［ 2025-12-14 ］. https：//arxiv.org/abs/2402.02392 https://arxiv.org/abs/2402.02392 .

Sun C N ， Huang S J ， Pompili D . LLM-based multi-agent decision-making： challenges and future directions ［J］. IEEE Robotics and Automation Letters ， 2025 ， 10 （ 6 ）： 5681 - 5688 .

Cheng Y H ， Zhao H ， Zhou X Y ， et al . A large language model for advanced power dispatch ［J］. Scientific Reports ， 2025 ， 15 ： 8925 .

Michelon F ， Zhou Y H ， Morstyn T . Large language model interface for home energy management systems ［C］// Proceedings of the 16th ACM International Conference on Future and Sustainable Energy Systems ， 2025 ： 590 - 602 .

Zhang B R ， Li C J ， Chen G ， et al . Large language model assisted optimal bidding of BESS in FCAS market： an AI-agent based approach ［PP/OL］. ［ 2025-12-14 ］. https：//arxiv.org/abs/2406.00974 https://arxiv.org/abs/2406.00974 .

Lou C W ， Jin Z K ， Tang W ， et al . LLM-enhanced multi-agent reinforcement learning with expert workflow for real-time P2P energy trading ［PP/OL］. ［ 2025-12-14 ］. https：//arxiv.org/abs/2507.14995 https://arxiv.org/abs/2507.14995 .

Guo T C ， Chen X Y ， Wang Y Q ， et al . Large language model based multi-agents： a survey of progress and challenges ［PP/OL］. ［ 2025-12-14 ］. https：//arxiv.org/abs/2402.01680 https://arxiv.org/abs/2402.01680 .

Feng X C ， Dou L X ， Li E ， et al . A survey on large language model-based social agents in game-theoretic scenarios ［PP/OL］. ［ 2025-12-14 ］. https：//arxiv.org/abs/2412.03920 https://arxiv.org/abs/2412.03920 .

Leng Y ， Yuan Y . Do LLM agents exhibit social behavior？［PP/OL］. ［ 2025-12-14 ］. https：//arxiv.org/abs/2312.15198 https://arxiv.org/abs/2312.15198 .

Zhang X M ， Chen Y X ， Yeh S ， et al . MetaMind： modeling human social thoughts with metacognitive multi-agent systems ［PP/OL］. ［ 2025-12-14 ］. https：//arxiv.org/abs/2505.18943 https://arxiv.org/abs/2505.18943 .

Zall R ， Kheyrkhah A ， Cambria E ， et al . Intelligent agents with emotional intelligence： current trends， challenges， and future prospects ［PP/OL］. ［ 2025-12-14 ］. https：//arxiv.org/abs/2511.20657 https://arxiv.org/abs/2511.20657 .

Tweedale J ， Ichalkaranje N ， Sioutis C ， et al . Innovations in multi-agent systems ［J］. Journal of Network and Computer Applications ， 2007 ， 30 （ 3 ）： 1089 - 1115 .

Jafari M ， Hua D Y ， Xue H ， et al . Enhancing conversational agents with theory of mind： aligning beliefs， desires， and intentions for human-like interaction ［PP/OL］. ［ 2025-12-14 ］. https：//arxiv.org/abs/2502.14171 https://arxiv.org/abs/2502.14171 .

Wei J ， Wang X Z ， Schuurmans D ， et al . Chain-of-thought prompting elicits reasoning in large language models ［PP/OL］. ［ 2025-12-14 ］. https：//arxiv.org/abs/2201.11903 https://arxiv.org/abs/2201.11903 .

Yao S Y ， Yu D ， Zhao J ， et al . Tree of thoughts： deliberate problem solving with large language models ［PP/OL］. ［ 2025-12-14 ］. https：//arxiv.org/abs/2305.10601 https://arxiv.org/abs/2305.10601 .

Kojima T ， Gu S S ， Reid M ， et al . Large language models are zero-shot reasoners ［PP/OL］. ［ 2025-12-14 ］. https：//arxiv.org/abs/2205.11916 https://arxiv.org/abs/2205.11916 .

Collins J ， Ketter W . Power TAC： software architecture for a competitive simulation of sustainable smart energy markets ［J］. SoftwareX ， 2022 ， 20 ： 101217 .

Bi X ， Chen D L ， Chen G T ， et al . DeepSeek LLM： scaling open-source language models with longtermism ［PP/OL］. ［ 2025-12-14 ］. https：//arxiv.org/abs/2401.02954 https://arxiv.org/abs/2401.02954 .

Yang A ， Li A F ， Yang B S ， et al . Qwen3 technical report ［PP/OL］. ［ 2025-12-14 ］. https：//arxiv.org/abs/2505.09388 https://arxiv.org/abs/2505.09388 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Local Electricity Market Trading Mechanism Based on Combinatorial Clock Auction

Decision-making Model of Carbon Emission Trading for Power Generators and Consumers Considering Electricity Spot Trading

Application and Prospect of Blockchain Technology in Electricity Trading

Review of Integration Mechanism of Data Centers and Energy Storage for Computing Power-Electric Power Coordination

Coordinated Optimal Scheduling of Computing Power-Electric Power Resources in Multi-data Centers Based on Accelerated Deep Reinforcement Learning

Related Author

JING Zhaoxia

PAN Jiajing

ZHAO Yuxuan

JI Tianyao

SHANG Nan

LENG Yuan

CHEN Zheng

WANG Kun

Related Institution

School of Electric Power Engineering, South China University of Technology

Energy Development Research Institute, China Southern Power Grid Co., Ltd.

Beijing Electric Power Trading Center

State Key Laboratory of Control and Simulation of Power System and Generation Equipments, Tsinghua University

Department of Distribution Network Planning Research, China Southern Power Grid Energy Development Research Institute

AI问答

⁰