| Issue |
Wuhan Univ. J. Nat. Sci.
Volume 31, Number 3, June 2026
|
|
|---|---|---|
| Page(s) | 225 - 240 | |
| DOI | https://doi.org/10.1051/wujns/2026313225 | |
| Published online | 24 June 2026 | |
Computer Science
CLC number: TM732
Analysis of Tripartite Evolutionary Game in Marketization of New Energy Electricity Prices Based on Large Language Models
基于大语言模型的新能源电价市场化三方演化博弈分析: 毛卉1,2,谭本艳1†,KHANAL Ribesh1,ULLAH Obaid1,黄美忠2
1
School of Economics and Management, China Three Gorges University, Yichang 443002, Hubei, China
2
School of Artificial Intelligence and Big Data, Wuhan Business University, Wuhan 430056, Hubei, China
† Corresponding author. E-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
Received:
28
December
2025
Abstract
The newly-issued 2025 policy on deepening the market-oriented reform of new energy feed-in tariffs has exerted a profound impact on reshaping the development pattern of new energy industries, such as photovoltaic power. In this evolving context, collaborative grid-connection among photovoltaic power generation enterprises, power grid enterprises, and government agencies is crucial for enhancing the competitiveness of the new energy industry and achieving energy transition. This paper constructs a tripartite evolutionary game model to deeply explore the strategy selection and key influencing factors of each subject in the grid-connection process. It integrates large language models (LLMs) to analyze factors affecting strategy selection among different stakeholders and utilizes LLMs to capture the heterogeneous cognitive characteristics of different subjects, thereby overcoming the limitations of "strong assumptions" commonly found in traditional game models. Through multi-round semantic parsing, it identifies key influencing factors such as market-oriented electricity price fluctuations, technological innovation costs, and assessment penalty. Furthermore, based on the actual data of photovoltaic industry development in Jiangxi and Hubei Provinces, numerical simulations are employed to analyze the impact of key factors (e.g., market-oriented electricity price fluctuations) on the strategic choices of the three stakeholder parties under the new policy framework and verify the model's effectiveness. The study clarifies the critical thresholds affecting collaborative grid connection, providing a data-driven theoretical basis for the government to implement targeted policies and enterprises to optimize decision-making.
摘要
新能源上网电价市场化改革对光伏等新能源产业发展格局带来了深刻影响,在此背景下,光伏发电企业、电网企业和政府三方协同上网对提升新能源产业竞争力、实现能源转型具有重要意义。本文构建三方演化博弈模型,深入探讨各主体在上网过程中的策略选择及影响因素。将大语言模型(LLMs)整合到影响各主体策略选择的因素分析中,利用LLMs捕捉不同主体异质性认知特征,突破传统博弈模型“强假设”局限,通过多轮语义解析筛选出市场化价格波动、技术创新成本、机制电量占比等核心影响因素。进一步,本文基于江西省与湖北省光伏产业发展的实际数据,采用数值模拟方法,分析新型政策框架下市场化电价波动等关键因素对三方利益主体策略选择的影响,并验证模型有效性;识别了影响协同上网的关键阈值,为政府精准施策、企业优化决策提供了数据驱动的理论依据。
Key words: electricity marketization in 2025 / large language models (LLMs) / policy simulation / tripartite evolutionary game
关键字 : 2025年电价市场化新政策 / LLMs / 政策仿真 / 三方演化博弈
Cite this article: MAO Hui, TAN Benyan, KHANAL Ribesh, et al. Analysis of Tripartite Evolutionary Game in Marketization of New Energy Electricity Prices Based on Large Language Models[J]. Wuhan Univ J of Nat Sci, 2026, 31(3): 225-240.
Biography: MAO Hui, female, Ph.D. candidate, research direction: AI technology application. E-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
Foundation item: Supported by the National Social Science Foundation of China (23BJY013)
© Wuhan University 2026
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.
