Machine Learning Based Distributionally Robust Optimization and Coordinated Operation of Electric Power, Computing Power, and Thermal Power in Data Centers

LIU Chengxi; WENG Zebin; LUO Yongjian; LI Zhuoqun; ZHANG Qian; HAN Jiangbei

doi:10.7500/AEPS20251117007

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

Machine Learning Based Distributionally Robust Optimization and Coordinated Operation of Electric Power, Computing Power, and Thermal Power in Data Centers

Computing Power-Electric Power Coordination | 更新时间：2026-10-10

- Machine Learning Based Distributionally Robust Optimization and Coordinated Operation of Electric Power, Computing Power, and Thermal Power in Data Centers
- Automation of Electric Power Systems Vol. 50, Issue 19, Pages: 87-95(2026)
- 作者机构：
  
  1.武汉大学电气及自动化学院，湖北省武汉市 430072
  2.国网北京市电力公司，北京市 100031
- 作者简介：
- 基金信息：
  
  Smart-Grid National Science and Technology Major Project(2025ZD0804800)
- DOI：10.7500/AEPS20251117007
  CLC：
- Received：17 November 2025，
  
  Revised：2025-12-08，
  
  Online First：25 May 2026，
  
  Published：10 October 2026
- 稿件说明：
移动端阅览
刘承锡,翁泽斌,罗永建等.基于机器学习的数据中心电-算-热分布鲁棒优化与协同运行[J].电力系统自动化,2026,50(19):87-95.

LIU Chengxi,WENG Zebin,LUO Yongjian,et al.Machine Learning Based Distributionally Robust Optimization and Coordinated Operation of Electric Power, Computing Power, and Thermal Power in Data Centers[J].Automation of Electric Power Systems,2026,50(19):87-95.
刘承锡,翁泽斌,罗永建等.基于机器学习的数据中心电-算-热分布鲁棒优化与协同运行[J].电力系统自动化,2026,50(19):87-95. DOI： 10.7500/AEPS20251117007.

LIU Chengxi,WENG Zebin,LUO Yongjian,et al.Machine Learning Based Distributionally Robust Optimization and Coordinated Operation of Electric Power, Computing Power, and Thermal Power in Data Centers[J].Automation of Electric Power Systems,2026,50(19):87-95. DOI： 10.7500/AEPS20251117007.

摘要

人工智能等技术的快速发展使数据中心算力需求激增，进而带来巨大的电能消耗。然而，刚性的算力需求及大电网供电模式使数据中心难以有效实现降碳减排与绿色经济运行。因此，充分挖掘电-算-热灵活性对提高新能源供电的数据中心电-算供需动态匹配及绿电消纳至关重要。文中首先基于机器学习方法量化表征新能源不确定性，设计两阶段电-算-热协同运行分布鲁棒优化框架，完成多能耦合建模；其次，引入批处理任务转移系数矩阵，对可控算力任务的柔性调节特性进行量化分析；最后，通过协同优化对影响数据中心经济性的关键指标进行研究。结果表明，多能协同运行能节约49.75%的成本，其中

仅调节算力灵活性能改善48.5%的经济性，表明算力任务灵活调节特性对经济运行至关重要。

Abstract

The rapid development of artificial intelligence and other technologies has led to a surge in computing power demand in data centers

which in turn results in enormous energy consumption. However

the rigid computing power demand and power supply modes of large-scale power grids make it difficult for data centers to effectively reduce carbon emission and achieve green economic operation. Therefore

fully exploiting the flexibility of electric power

computing power and thermal power is crucial for improving the dynamic matching between electricity supply and computing demand

as well as enhancing the accommodation of renewable energy in data centers powered by green electricity. This paper first quantifies and characterizes the uncertainty of renewable energy based on machine learning methods

designs a two-stage distributionally robust optimization framework for coordinated operation of electric power

computing power and thermal power

and establishes multi-energy coupling modeling. Second

a batch task transfer coefficient matrix is introduced to quantitatively analyze the flexible regulation characteristics of controllable computing tasks. Finally

key indicators affecting the economic performance of data centers are investigated through coordinated optimization. The results of the case study show that the multi-energy coordinated operation can reduce costs by 49.75%

of which regulating computing power flexibility alone improves economic performance by 48.5%

indicating that the flexible regulation characteristics of computing power tasks are crucial for economic operation

关键词

Keywords

references

Masanet E ， Shehabi A ， Lei N ， et al . Recalibrating global data center energy-use estimates ［J］. Science ， 2020 ， 367 ： 984 - 986 .

Ajayi O ， Heymann R . Data centre day-ahead energy demand prediction and energy dispatch with solar PV integration ［J］. Energy Reports ， 2021 ， 7 ： 3760 - 3774 .

Koot M ， Wijnhoven F . Usage impact on data center electricity needs： a system dynamic forecasting model ［J］. Applied Energy ， 2021 ， 291 ： 116798 .

中国信通院 . 2025综合算力指数报告发布［EB/OL］.［ 2025-11-17 ］. https：//www.dt.gov.cn/dtszf/sxyw/202508/3b4f8738ba5f45dc94c724c9be46c0f3.shtml https://www.dt.gov.cn/dtszf/sxyw/202508/3b4f8738ba5f45dc94c724c9be46c0f3.shtml .

China Academy of Information and Communications Technology . 2025 Comprehensive computing power index report released ［EB/OL］. ［ 2025-11-17 ］. https：//www.dt.gov.cn/dtszf/sxyw/202508/3b4f8738ba5f45dc94c724c9be46c0f3.shtml https://www.dt.gov.cn/dtszf/sxyw/202508/3b4f8738ba5f45dc94c724c9be46c0f3.shtml .

杨挺，姜含，侯昱丞，等 . 基于计算负荷时-空双维迁移的互联多数据中心碳中和调控方法研究［J］. 中国电机工程学报， 2022 ， 42 （ 1 ）： 164 - 177 .

Yang Ting ， Jiang Han ， Hou Yucheng ， et al . Carbon-neutral regulation method for interconnected multi-data centers based on spatiotemporal two-dimensional migration of computing load ［J］. Proceedings of the CSEE ， 2022 ， 42 （ 1 ）： 164 - 177 .

俞露杰，周欢，朱介北，等 . 面向新能源消纳的数据中心热力学模型和变温运行策略［J/OL］. 电力系统自动化， 2025-11-17 . http：//doi. org/10.7500/AEPS20240414005 http://doi.org/10.7500/AEPS20240414005 .

Yu Lujie ， Zhou Huan ， Zhu Jiebei ， et al . Thermodynamic model and variable-temperature operation strategy of data centers for renewable energy accommodation ［J/OL］. Automation of Electric Power Systems ， 2025-11-17 . http：//doi. org/10.7500/AEPS20240414005 http://doi.org/10.7500/AEPS20240414005 .

Ren X ， Han Z ， Ma J ， et al . Life-cycle-based multi-objective optimal design and analysis of distributed multi-energy systems for datacenters ［J］. Energy ， 2024 ， 288 ： 129679 .

Han J ， Han K ， Han T ， et al . Data-driven distributionally robust optimization of low-carbon data center energy systems considering multi-task response and renewable energy uncertainty ［J］. Journal of Building Engineering ， 2025 ， 102 ： 111937 .

王天琪，于浩，赵金利，等 . 算力-热力灵活性协同的数据中心能量管理方法［J］. 高电压技术， 2024 ， 50 （ 9 ）： 4069 - 4079 .

Wang Tianqi ， Yu Hao ， Zhao Jinli ， et al . Data center energy management method with coordinated flexibility of computing power and thermal power ［J］. High Voltage Engineering ， 2024 ， 50 （ 9 ）： 4069 - 4079 .

Yang Z ， Trivedi A ， Liu H ， et al . Two-stage robust optimization strategy for spatially-temporally correlated data centers with data-driven uncertainty sets ［J］. Electric Power Systems Research ， 2023 ， 221 ： 109443 .

Kwon S . Ensuring renewable energy utilization with quality of service guarantee for energy-efficient data center operations ［J］. Applied Energy ， 2020 ， 276 ： 115424 .

Fan W ， Ju L ， Tan Z ， et al . Two-stage distributionally robust optimization model of integrated energy system group considering energy sharing and carbon transfer ［J］. Applied Energy ， 2023 ， 331 ： 120426 .

Taghizad-Tavana K ， Tarafdar-Hagh M ， Nojavan S ， et al . Operation of smart distribution networks by considering the spatial-temporal flexibility of data centers and battery energy storage systems ［J］. Sustainable Cities and Society ， 2024 ， 114 ： 105746 .

Zhou Y ， Li X ， Han H ， et al . Resilience-oriented planning of integrated electricity and heat systems： a stochastic distributionally robust optimization approach ［J］. Applied Energy ， 2024 ， 353 ： 122053 .

Zhao B ， Qian T ， Tang W ， et al . A data-enhanced distributionally robust optimization method for economic dispatch of integrated electricity and natural gas systems with wind uncertainty ［J］. Energy ， 2022 ， 243 ： 123113 .

Li W ， Qian T ， Zhang Y ， et al . Distributionally robust chance-constrained planning for regional integrated electricity-heat systems with data centers considering wind power uncertainty ［J］. Applied Energy ， 2023 ， 336 ： 120787 .

Zhang S ， Lyu J ， Jin W ， et al . Coordinated planning of multiple energy hubs considering the spatiotemporal load regulation of data centers ［J］. IEEE Transactions on Power Systems ， 2024 ， 39 （ 2 ）： 4193 - 4207 .

尹卫方，彭淑芬，孙毅，等 . 边缘数据中心集群算力-电力-热力协同优化运行策略［J/OL］. 电力信息与通信技术， 2025-11-17 . https：//link.cnki.net/urlid/10.1164.tk.20250527.1405.002 https://link.cnki.net/urlid/10.1164.tk.20250527.1405.002 .

Yin Weifang ， Peng Shufen ， Sun Yi ， et al . Coordinated optimal operation strategy of computing-electric-thermal power for edge data center clusters ［J/OL］. Electric Power Information and Communication Technology ， 2025-11-17 . https：//link.cnki.net/urlid/10.1164.tk.20250527.1405.002 https://link.cnki.net/urlid/10.1164.tk.20250527.1405.002 .

Zhou Y ， Wei F ， Li S ， et al . Data center load modeling through optimal energy consumption characteristics： a path to simultaneously enhance energy efficiency and demand response quality ［J］. Applied Energy ， 2025 ， 393 ： 126095 .

Wang Y ， Lin J ， Han Y ， et al . Comprehensive evaluation of all-element flexibility resources in data centers： considering synergistic benefits of computing， electricity， and heat ［J］. Applied Energy ， 2025 ， 399 ： 126442

国家发展和改革委员会 . 数据中心绿色低碳发展专项行动计划［EB/OL］.［ 2025-11-17 ］. https：//www.gov.cn/zhengce/zhengceku/202407/content_6964165.htm https://www.gov.cn/zhengce/zhengceku/202407/content_6964165.htm .

National Development and Reform Commission . Special action plan for the green and low-carbon development of data centers ［EB/OL］. ［ 2025-11-17 ］. https：//www.gov.cn/zhengce/zhengceku/2024-07/content_6964165.htm https://www.gov.cn/zhengce/zhengceku/2024-07/content_6964165.htm .

Rasmussen N . Electrical efficiency modeling for data centers ［EB/OL］. ［ 2025-11-17 ］. https：//it-resource.schneider-electric.com/whitepapers/wp-113-electrical-efficiency-modeling-for-data-centers https://it-resource.schneider-electric.com/whitepapers/wp-113-electrical-efficiency-modeling-for-data-centers .

羲和能源气象大数据平台［E B/OL］.［ 2025-11-17 ］. https：//xihe-energy.com/ https://xihe-energy.com/ .

Xihe energy meteorological big data platform ［EB/OL］. ［ 2025-11-17 ］. https：//xihe-energy.com/ https://xihe-energy.com/ .

Google . Traffic and disruptions to Google ［EB/OL］. ［ 2025-11-17 ］. https：//transparencyreport.google.com/traffic/overview？fraction_traffic=start：1722384000000；end：1722556799999；product：19；region：HK&lu=fraction_traffic&hl=en https://transparencyreport.google.com/traffic/overview?fraction_traffic=start:1722384000000;end:1722556799999;product:19;region:HK&lu=fraction_traffic&hl=en .

Weng Q ， Xiao W ， Yu Y ， et al . Mlaas in the wild： workload analysis and scheduling in large-scale heterogeneous gpu clusters ［EB/OL］. ［ 2025-11-17 ］. https：//web.stanford.edu/class/cs349d/docs/MLaaS_in_the_Wild.pdf https://web.stanford.edu/class/cs349d/docs/MLaaS_in_the_Wild.pdf .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Computing Power-Electric Power Coordinated Dispatch Method for Wide-area Data Center Clusters Based on Spatial Flexibility Characterization

Green Power Directly-connected Data Centers: Coupling Logic, Key Technologies, and Application Modes

Optimal Allocation Method for Centralized Backup Energy Storage of Data Center Clusters Under Shared Framework

Optimization Method for Coordinated Operation of Computing Power Network and Electric Power Network Under Multi-source Uncertainty Risk Constraints

Method and Practice of Computing Power-Electric Power Coordination Based on Idle Computing Power Scheduling

Related Author

FENG Hongyun

TAN Zhenfei

YAN Zheng

DING Hao

TONG Jie

YI Chenle

WANG Zhanbo

CHEN Hongyin

Related Institution

School of Electrical Engineering, Shanghai Jiaotong University

School of Electrical and Electronic Engineering, North China Electric Power University

China Electric Power Research Institute

College of Electrical Engineering, Zhejiang University

Guizhou Power Grid Co., Ltd.

AI问答

⁰