主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2024, Vol. 32 ›› Issue (12): 312-322.doi: 10.16381/j.cnki.issn1003-207x.2023.1944

• • 上一篇    下一篇

基于主从博弈的虚拟电厂实时定价需求响应机制研究

解百臣1, 章露1, 郝鹏2()   

  1. 1.天津大学管理与经济学部,天津 300072
    2.天津仁爱学院经济与管理学院,天津 301636
  • 收稿日期:2023-11-08 修回日期:2024-02-05 出版日期:2024-12-25 发布日期:2025-01-02
  • 通讯作者: 郝鹏 E-mail:phao@tju.edu.cn
  • 基金资助:
    国家社会科学基金重大项目(22&ZD104);国家自然科学基金项目(72243009)

Real-time Pricing Demand Response Mechanism of Virtual Power Plant Based on Stackelberg Game

Baichen Xie1, Lu Zhang1, Peng Hao2()   

  1. 1.College of Economics and Management,Tianjin University,Tianjin 300072,China
    2.Economics and Management College,Tianjin Ren-ai College,Tianjin 301636,China
  • Received:2023-11-08 Revised:2024-02-05 Online:2024-12-25 Published:2025-01-02
  • Contact: Peng Hao E-mail:phao@tju.edu.cn

摘要:

虚拟电厂通过运营商聚合分布式能源、可调负荷和储能系统,可以有效调控负荷用电行为,提高需求侧管理效率。本文考虑虚拟电厂运营商与产消者、外部主网的双向电力交易,以储能充放电量的日前优化结果为输入,基于主从博弈模型研究虚拟电厂实时定价的需求响应机制。本研究以主网、虚拟电厂运营商、终端负荷的协同调度为目标,采用分布式遗传算法求解各博弈主体的均衡策略。结果表明,实时定价的需求响应机制能有效调控用户用电行为,助力虚拟电厂实现削峰填谷、平抑负荷波动的目标;均衡状态下,不同类型用户经济性存在显著差异:配备了燃气轮机发电的光伏产消者,参与需求响应的效益降低;单独配备燃气轮机发电的产消者,能通过参与需求响应减少效益损失;无发电能力的用户、光伏产消者及虚拟电厂运营商,均能通过参与需求响应实现效益提升。综合以上分析,文章从挖掘资源潜力、接入辅助服务市场、网厂互联等角度提出了未来加强虚拟电厂建设的建议。

关键词: 需求响应, 虚拟电厂, 主从博弈, 实时定价

Abstract:

Distributed energy plays an important role in mitigating climate change. Its large-scale application impacts the stability of the power system due to the intermittent property. Virtual power plants (VPPs), which aggregate distributed energy resources, adjustable loads, and energy storage systems, can effectively regulate load consumption and promote renewable energy consumption. We studied the demand response (DR) mechanism of real-time pricing within the VPP based on the Stackelberg game model, considering two-way electricity transactions between the VPP operator and prosumers, as well as between the VPP operator and the external main grid (MG). The study focuses on the collaborative scheduling of the MG, the VPP operator, and terminal loads. A distributed genetic algorithm is used to solve the equilibrium strategy for each stakeholder. The numerical simulation results show that: (1) The real-time pricing-based DR mechanism effectively incentivizes the VPP operator to trade with internal loads, smoothing load fluctuations and achieving significant peak-shaving and valley-filling effects. (2) From the users’perspective, both non-generating users and PV prosumers can benefit from participating in DR, while prosumers with only gas turbines have limited overall benefits. (3) The introduction of a real-time pricing DR mechanism effectively improves the benefits of VPPs and facilitates power resource scheduling. Based on these findings, the paper offers three suggestions for strengthening the future development of VPPs: First, VPPs should fully tap the controllable potential of loads, energy storage and other resources, guiding the loads to participate in peak shaving and valley filling through price signals; Second, to incentivize PV prosumers equipped with gas turbines to participate in DR, their opportunities to participate in the ancillary service market should be broadened, maximizing the income from peak balancing potential; Finally, an MG-VPP interconnection system based on artificial intelligence and big data should be developed. This system would integrate the MG’s real-time price and load forecasting information with the VPP’s functions of energy production, energy storage, and demand response.

Key words: demand response, virtual power plant, stackelberg game, real-time pricing

中图分类号: