综合智慧能源 ›› 2026, Vol. 48 ›› Issue (3): 15-26.doi: 10.3969/j.issn.2097-0706.2026.03.002

• 电力系统建模与调控 • 上一篇    下一篇

基于多智能体强化学习的多能互补能源系统优化运行

陈锋1,2(), 路小敏3(), 李梦杨4(), 张涛1,2(), 杨帆1,2()   

  1. 1 河南科技大学 应用工程学院河南 三门峡 472100
    2 三门峡职业技术学院 河南省有色金属新材料智能制造应用工程研究中心河南 三门峡 472099
    3 郑州浪潮数据技术有限公司郑州 450003
    4 洛阳师范学院 电气工程与自动化学院河南 洛阳 471942
  • 收稿日期:2025-07-18 修回日期:2025-12-08 出版日期:2026-03-25
  • 作者简介:陈锋(1987),男,副教授,高级工程师,硕士,从事高比例新能源新型电力系统优化运行与智能控制等方面的研究,chenfeng@smxpt.edu.cn
    路小敏(1989),女,工程师,硕士,从事电力大数据技术等方面的研究,703399594@qq.com
    李梦杨(1988),女,副教授,博士,从事非线性系统鲁棒控制、多智能体、工业智能、无人驾驶等方面的研究,limengyang8801@163.com
    张涛(1995),男,工程师,硕士,从事电力电子与电力传动等方面的研究,474872683@qq.com
    杨帆(1989),女,讲师,硕士,从事跨区域电力联网背景下人力资源统筹配置等方面的研究,18239861350@163.com
  • 基金资助:
    国家自然科学基金项目(62401240);河南省教育厅高等学校重点科研项目计划(26B480004)

Optimization operation of multi-energy complementary system based on multi-agent reinforcement learning

CHEN Feng1,2(), LU Xiaomin3(), LI Mengyang4(), ZHANG Tao1,2(), YANG Fan1,2()   

  1. 1 School of Applied EngineeringHenan University of Science and TechnologySanmenxia 472100, China
    2 Henan Nonferrous Metals New Materials Intelligent Manufacturing Application Engineering Research CenterSanmenxia PolytechnicSanmenxia 472099, China
    3 Inspur Data Technology Company LimitedZhengzhou 450003, China
    4 School of Electrical Engineering and AutomationLuoyang Normal UniversityLuoyang 471942, China
  • Received:2025-07-18 Revised:2025-12-08 Published:2026-03-25
  • Supported by:
    National Natural Science Foundation of China(62401240); Key Scientific Research Project Plan of Colleges and Universities in Henan Province in 2026(26B480004)

摘要:

针对多能互补能源系统在高比例可再生能源接入下的动态协同优化难题,以及传统集中式方法在多主体利益协调和实时响应中的局限性,开展动态优化建模研究。构建了“物理层-决策层-协同层”三层多智能体强化学习(MARL)框架,将能源生产者、消费者及系统调度器划分为独立智能体。基于改进近端策略优化算法,设计了融合经济性、环保性与稳定性的动态奖励函数,通过集中训练-分散执行机制实现分布式决策与全局协同。以典型的园区级多能互补系统为算例,结果表明:所提MARL模型使可再生能源消纳率提升至92.3%,单位电量成本较传统混合整数规划(MIP)方法降低了28.9%;在50%负荷突变场景下,系统恢复稳定时间缩短至90 s,较MIP方法提速900%;面对±20%风光预测误差,负荷满足率仍保持98.7%。该动态优化模型可有效解决多能互补系统的多主体协同与不确定性适应问题,为高渗透率可再生能源系统的实时优化调度提供技术支撑。

关键词: 多能互补能源系统, 多智能体强化学习, 动态优化建模, 源网荷储, 可再生能源消纳, 协同调度

Abstract:

To address the dynamic coordinated optimization challenges of multi-energy complementary systems under high renewable energy integration and the limitations of traditional centralized methods in multi-agent interest coordination and real-time response, dynamic optimization modeling research was conducted. A three-layer multi-agent reinforcement learning (MARL) framework—consisting of a physical layer, decision layer, and coordination layer—was developed, with energy producers, consumers, and system schedulers classified as independent agents. Based on the improved proximal policy optimization algorithm, a dynamic reward function integrating economic efficiency, environmental friendliness, and stability was designed, and distributed decision-making with global coordination was achieved through a centralized training-decentralized execution mechanism. A typical park-level multi-energy complementary system was used as a case study. The results showed that the proposed MARL model increased the renewable energy consumption rate to 92.3%, reducing the unit electricity cost by 28.9% compared to the traditional mixed integer programming (MIP) method. Under a 50% load abrupt change scenario, the system recovery time was shortened to 90 s, which was 900% faster than the MIP method. Even with ±20% wind and solar forecasting errors, the load satisfaction rate remained at 98.7%. This dynamic optimization model effectively addressed the multi-agent coordination and uncertainty adaptation challenges in multi-energy complementary systems, providing technical support for the real-time optimization and scheduling of high-penetration renewable energy systems.

Key words: multi-energy complementary system, multi-agent reinforcement learning, dynamic optimization modeling, source-grid-load-storage, renewable energy consumption, coordinated scheduling

中图分类号: