基于多场景下的考虑CVaR和碳交易成本的虚拟电厂优化调度策略

Optimized scheduling strategy for virtual power plant considering CVaR and carbon trading costs under multiple scenarios

ES评分 0

DOI 10.12208/j.jeea.20250004
刊名
Journal of Electrical Engineering and Automation
年,卷(期) 2025, 4(1)
作者
作者单位

1 南京邮电大学自动化学院/人工智能学院 江苏南京 2 国网常州供电公司设备监控班 江苏常州

摘要
本文研究了在多市场环境下,结合条件风险价值(CVaR)和碳成本的虚拟电厂(VPP)优化调度策略。虚拟电厂由分布式燃气轮机、储能设备、风力发电、光伏发电及非柔性负荷组成,研究的核心目标是在基于多场景下VPP通过参与电力期货市场、日前市场、实时市场以及碳市场交易,分析最大化收益下的VPP内部优化调度结果。在实际操作中,当日前市场的风电和光伏发电无法满足非柔性负荷需求时,VPP优先满足期货市场的负荷需求,并通过与实时市场的博弈,决定是购电还是启动燃气轮机发电,以平衡供需。本文详细分析了多种风电、光伏出力及负荷场景下的优化调度结果,旨在为决策者在极端天气条件下提供科学的决策支持。同时为了应对风电和光伏出力的波动带来的不确定性,本文采用蒙特卡罗模拟结合CVaR方法,对出力波动的不确定性进行量化,并将其视为相关成本加以考虑。通过分析不同偏差容忍度下的调度优化结果,研究表明,合理设置偏差容忍度不仅能够在多市场环境中实现较高收益,还能有效降低运营风险,从而推动构建更加低碳和高效的能源系统。
Abstract
This paper investigates the optimization scheduling strategy of Virtual Power Plant in a multi-market environment, integrating Conditional Value-at-Risk (CVaR) and carbon cost considerations. The VPP comprises distributed gas turbines, energy storage devices, wind power, photovoltaic generation, and non-flexible loads. The core objective of this research is to analyze the internal optimization scheduling results of VPPs that maximize profits through participation in electricity futures markets, day-ahead markets, real-time markets, and carbon markets under a multi-scenario approach. In practical operations, when wind and photovoltaic generation in the day-ahead market cannot meet non-flexible load demands, the VPP prioritizes satisfying load demands in the futures market. Through strategic interactions with the real-time market, the VPP determines whether to purchase electricity or activate gas turbines for power generation to balance supply and demand. This paper presents a detailed analysis of optimization scheduling results under various wind power, photovoltaic output, and load scenarios, aiming to provide scientific decision support for decision-makers under extreme weather conditions. Furthermore, to address the uncertainties caused by fluctuations in wind and photovoltaic outputs, this study employs Monte Carlo simulation combined with the CVaR method to quantify output volatility, which is considered as a relevant cost factor. By analyzing optimization results under different deviation tolerance levels, the research demonstrates that setting reasonable deviation tolerance not only achieves higher returns in multi-market environments but also effectively reduces operational risks, thereby promoting the development of a more low-carbon and efficient energy system.
关键词
虚拟电厂;碳成本;多市场环境;不确定性;条件风险价值;偏差容忍度;多场景;蒙特卡罗模拟
KeyWord
Virtual Power Plant; Carbon Cost; Multi-Market Environment; Uncertainty; Conditional Value at Risk (CVaR); Deviation Tolerance; Multi-Scenario; Monte Carlo Simulation
基金项目
页码 21-40
  • 参考文献
  • 相关文献
  • 引用本文

葛辉, 龚蕾, 过祈睿, 刘程子, 刘泽远, 杨艳. 基于多场景下的考虑CVaR和碳交易成本的虚拟电厂优化调度策略 [J]. 电气工程与自动化. 2025; 4; (1). 21 - 40.

  • 文献评论

相关学者

相关机构