| 引用本文: | 崔杨,程禹烽,赵钰婷,李佳宇,李崇钢.考虑特性分类批处理负荷可调节能力的数据中心微网灵活性设备分布鲁棒容量配置方法[J].电力自动化设备,2024,44(7):180-188 |
| CUI Yang,CHENG Yufeng,ZHAO Yuting,LI Jiayu,LI Chonggang.Distributionally robust capacity allocation method for flexibility device of data center microgrid considering adjustable capability of characteristic classification batch processing loads[J].Electric Power Automation Equipment,2024,44(7):180-188 |
|
| 摘要: |
| 微网灵活性设备主要用于平抑源荷两侧的波动，其容量配置方法应考虑源荷不确定性的影响，而含数据中心微网的灵活性设备容量配置方法还应进一步考虑数据中心负荷的可调节特性。考虑数据中心批处理负荷的可调节能力和源荷不确定性因素，提出一种灵活性设备容量配置方法。根据负荷特性的不同，将批处理负荷划分为2类以更加准确地量化其可调节能力，一类为带宽时序可变限时可平移负荷，另一类为带宽时序不变可中断平移负荷，对这2类批处理负荷进行详细分析并给出一般性的建模方法；构建数据驱动下的min-max-min两阶段分布鲁棒优化容量配置模型，利用1-范数和∞-范数约束场景概率分布模糊集，采用列和约束生成算法对该模型进行化简求解。对某省数据中心微网进行算例分析，验证了所提方法的有效性。 |
| 关键词: 数据中心 批处理负荷 特性分类 可调节能力 容量配置 分布鲁棒优化 |
| DOI：10.16081/j.epae.202405012 |
| 分类号:TM715 |
| 基金项目:国家自然科学基金资助项目(51777027) |
|
| Distributionally robust capacity allocation method for flexibility device of data center microgrid considering adjustable capability of characteristic classification batch processing loads |
|
CUI Yang1, CHENG Yufeng1, ZHAO Yuting1, LI Jiayu2, LI Chonggang3
|
|
1.Key Laboratory of Modern Power System Simulation and Control & Renewable Energy Technology, Ministry of Education, Northeast Electric Power University, Jilin 132012, China;2.Luohe Power Supply Company of State Grid Henan Electric Power Company, Luohe 462000, China;3.Yantai Power Supply Company of State Grid Shandong Electric Power Company, Yantai 265200, China
|
| Abstract: |
| The microgrid flexibility devices are primarily used to mitigate the fluctuation on both source and load sides, the capacity allocation method should consider the impact of source and load uncertainty, while the capacity allocation method for microgrid containing data center should further consider the adjustable characteristic of data center loads. Considering the adjustable capacity of batch processing loads in data center and the uncertainty of source and load, a capacity allocation method for flexibility devices is proposed. According to different load characteristics, the batch processing loads are divided into two categories to accurately quantify their adjustability, one category includes bandwidth-variant, time-limited and shiftable loads, while the other category includes bandwidth-invariant, interruptible and shiftable loads. The two types of batch processing loads are detailedly analyzed and a general modeling method is given. A data-driven min-max-min two-stage distributionally robust optimization capacity allocation model is constructed, The 1-norm and ∞-norm constrained scenario probability distribution fuzzy sets are used and the column-and-constraint generation algorithm is adopted to simplify and solve the model. The example analysis is carried out with a provincial data center microgrid, which verifies the effectiveness of the proposed method. |
| Key words: data center batch processing loads characteristic classification adjustable capability capacity allocation distributionally robust optimization |
