电工技术学报  2021, Vol. 36 Issue (7): 1446-1455    DOI: 10.19595/j.cnki.1000-6753.tces.200929
新能源电力系统及装备 |
基于模糊-概率策略实时反馈的虚拟电厂多时间尺度优化调度
李翔宇, 赵冬梅
华北电力大学电气与电子工程学院 北京 102206
Research on Multi-Time Scale Optimal Scheduling of Virtual Power Plant Based on Real-Time Feedback of Fuzzy-Probability Strategy
Li Xiangyu, Zhao Dongmei
School of Electrical and Electronic Engineering North China Electric Power University Beijing 102206 China
全文: PDF (1585 KB)   HTML
输出: BibTeX | EndNote (RIS)      
摘要 虚拟电厂(VPP)日前计划和日内运行方案是其在内、外部随机性上,实现经济或技术效益的有力保证。提出VPP日前计划-滚动计划-实时调度在内的全时域优化调度框架,建立不同尺度优化模型并通过“多级调度、逐级细化”的思想达到优化鲁棒性。在日前计划中,以虚拟电厂最大化市场效益为目标建立模型;在日内优化中,基于马尔科夫决策过程(MDP)融合30min尺度滚动和5min尺度超短期两层优化以减少建模层级。在MDP模型中,建立熵值回馈函数表征滚动优化过程,用以减少因日前预测误差带来的日内运行偏差值;建立模糊-概率策略表征超短期优化过程,用以调节储能荷电状态、均衡其充电与放电能力,最终实现跟踪日前计划的目标,最大化经济效益。最后通过算例计算对所提方案进行了验证。
服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
李翔宇
赵冬梅
关键词 虚拟电厂优化调度多时间尺度马尔科夫决策过程模糊-概率策略    
Abstract:The day-ahead plan and operation scheme of virtual power plant (VPP) are the strong guarantee for realizing economic or technical benefits under its internal and external randomness. The multi-level optimization scheduling framework of VPP including day-ahead plan, rolling plan and real-time scheduling is proposed, and optimization models of different scales are established to achieve optimization robustness through the idea of ‘multi-level scheduling and step-by-step refinement’. In the day-ahead plan, a model is established with the goal of maximizing market benefits of virtual power plants. In intra-day optimization, based on the Markov decision process (MDP), the two-layer optimization of 30-min scale rolling plan and 5-min scale real-time scheduling are combined to reduce the modeling level. In the MDP model, the entropy feedback function is established to characterize the rolling optimization process, and the fuzzy-probability strategy is established to characterize the real-time optimization process to adjust the state of charge of energy storage (state of charge, SOC), which are finally effectively track the goals of day-ahead plan to maximize economic benefits. Finally, the proposed scheme was verified by calculation examples.
Key wordsVirtual power plan (VPP)    optimal scheduling    multi-time scale    Markov decision process (MDP)    fuzzy-probability strategy   
收稿日期: 2020-07-31     
PACS: TM62  
  TM73  
基金资助:国家重点研发计划资助项目(2017YFB0902600)
通讯作者: 李翔宇 男,1988年生,博士研究生,研究方向为新能源并网运行及其控制。E-mail:410489024@qq.com   
作者简介: 赵冬梅 女,1968年生,教授,博士生导师,研究方向为电力系统分析、稳定和控制、电力市场及新能源并网运行。E-mail:zhao-dm@ncepu. edu. cn
引用本文:   
李翔宇, 赵冬梅. 基于模糊-概率策略实时反馈的虚拟电厂多时间尺度优化调度[J]. 电工技术学报, 2021, 36(7): 1446-1455. Li Xiangyu, Zhao Dongmei. Research on Multi-Time Scale Optimal Scheduling of Virtual Power Plant Based on Real-Time Feedback of Fuzzy-Probability Strategy. Transactions of China Electrotechnical Society, 2021, 36(7): 1446-1455.
链接本文:  
https://dgjsxb.ces-transaction.com/CN/10.19595/j.cnki.1000-6753.tces.200929          https://dgjsxb.ces-transaction.com/CN/Y2021/V36/I7/1446