1.福州大学电气工程与自动化学院,福建省福州市 350116
2.数字能源福建省高校重点实验室(福州大学),福建省福州市 350116
庞奥康(1998—),男,博士研究生,主要研究方向:数据中心优化运行。E-mail:pakdeyouxiang@163.com
邵振国(1970—),男,博士,教授,博士生导师,主要研究方向:能源系统不确定理论与方法、电能质量分析与治理、电力大数据理论与方法等。E-mail:shao.zg@fzu.edu.cn
陈昌铭(1997—),男,通信作者,博士,校聘副研究员,硕士生导师,主要研究方向:多能数据中心优化、电力系统恢复与防灾应急。E-mail:chenchangming@fzu.edu.cn
收稿:2025-09-26,
修回:2026-01-28,
网络首发:2026-07-09,
纸质出版:2026-09-10
移动端阅览
庞奥康,邵振国,陈昌铭等.计及任务时序灵活性的数据中心算力-电力协同优化方法[J].电力系统自动化,2026,50(17):115-125.
PANG Aokang,SHAO Zhenguo,CHEN Changming,et al.Coordinated Optimization Method for Computing Power and Electric Power in Data Centers Considering Sequential Flexibility of Tasks[J].Automation of Electric Power Systems,2026,50(17):115-125.
庞奥康,邵振国,陈昌铭等.计及任务时序灵活性的数据中心算力-电力协同优化方法[J].电力系统自动化,2026,50(17):115-125. DOI: 10.7500/AEPS20250926005.
PANG Aokang,SHAO Zhenguo,CHEN Changming,et al.Coordinated Optimization Method for Computing Power and Electric Power in Data Centers Considering Sequential Flexibility of Tasks[J].Automation of Electric Power Systems,2026,50(17):115-125. DOI: 10.7500/AEPS20250926005.
随着人工智能技术的快速发展和算力需求的日益增加,数据中心能耗急剧上升,其高运营成本和高排放已成为制约其发展的瓶颈之一。为此,文中提出一种计及算力任务时序灵活性的数据中心算力-电力协同优化方法。首先,通过分析算力任务间的依赖特性,提出了基于有向无环图的算力任务序列分解模型,以精确量化各类算力任务的可延迟时间。其次,融合Transformer和柯尔莫哥洛夫-阿诺德神经网络建立算力任务-能耗映射模型,以精确表征室内外温度变化下算力任务与电/冷能耗间的非线性关系。最后,以最小化运营成本为目标,构建数据中心算力-电力协同优化模型。某公开数据集的验证结果表明,所提方法具备契合实际场景的算力任务灵活时移能力,能够有效降低系统总运营成本。
With the rapid development of artificial intelligence technologies and the increasing demand for computing power
the energy consumption of data centers has surged dramatically
and high operating costs and emissions have become one of the bottlenecks restricting their development. Therefore
a coordinated optimization method for computing power and electric power in data centers considering sequential flexibility of computing power tasks is proposed. Firstly
by analyzing the dependency characteristics among computing power tasks
a decomposition model for computing power task sequences based on directed acyclic graph is proposed to precisely quantify the deferrable time for various computing power tasks. Secondly
by integrating Transformer and Kolmogorov-Arnold network
a computing power task-energy consumption mapping model is built to accurately characterize the nonlinear relationship between computing power task and electricity/cooling energy consumption under varying indoor and outdoor temperatures. Finally
with the objective of minimizing operating costs
a coordinated optimization model for computing power and electric power in data centers is constructed. Validation results of a publicly available dataset indicate that the proposed method achieves flexible time-shifting capabilities for computing power tasks suitable for practical scenarios
and can effectively reduce the total operating costs of the system.
刘祎泽 , 向月 . 计及负荷转移需求响应的低碳数据中心光储容量优化配置 [J]. 电力自动化设备 , 2024 , 44 ( 7 ): 149 - 155 .
Liu Yize , Xiang Yue . Optimal allocation of photovoltaic and storage capacity in low carbon data center considering load transfer demand response [J]. Electric Power Automation Equipment , 2024 , 44 ( 7 ): 149 - 155 .
陈晓红 , 曹廖滢 , 陈姣龙 , 等 . 我国算力发展的需求、电力能耗及绿色低碳转型对策 [J]. 中国科学院院刊 , 2024 , 39 ( 3 ): 528 - 539 .
Chen Xiaohong , Cao Liaoying , Chen Jiaolong , et al . Development demand, power energy consumption and green and low-carbon transition for computing power in China [J]. Bulletin of Chinese Academy of Sciences , 2024 , 39 ( 3 ): 528 - 539 .
Zhang Z C , Zhang X , Bao Z X , et al . Courier: a unified communication agent to support concurrent flow scheduling in cluster computing [J]. IEEE Transactions on Parallel and Distributed Systems , 2025 , 36 ( 5 ): 861 - 876 .
Wang J J , Deng H D , Liu Y , et al . Coordinated optimal scheduling of integrated energy system for data center based on computing load shifting [J]. Energy , 2023 , 267 : 126585 .
Fan J Q , Yan R J , He Y , et al . Stochastic optimization of combined energy and computation task scheduling strategies of hybrid system with multi-energy storage system and data center [J]. Renewable Energy , 2025 , 242 : 122466 .
文亚凤 , 程祖铭 , 刘欣雅 , 等 . 考虑算力-电热耦合的边缘数据中心能量优化管理 [J]. 中国电机工程学报 , 2026 , 46 ( 1 ): 186 - 197 .
Wen Yafeng , Cheng Zuming , Liu Xinya , et al . Energy optimization management of edge data centers considering computing power-electrothermal coupling [J]. Proceedings of the CSEE , 2026 , 46 ( 1 ): 186 - 197 .
祁兵 , 曹望璋 , 李彬 , 等 . 计及负载特征及响应特性的多数据中心双层优化模型 [J]. 电力系统自动化 , 2022 , 46 ( 21 ): 30 - 41 .
Qi Bing , Cao Wangzhang , Li Bin , et al . Bi-level optimization model of multiple data centers considering workload features and response characteristics [J]. Automation of Electric Power Systems , 2022 , 46 ( 21 ): 30 - 41 .
文淅宇 , 朱继忠 , 李盛林 , 等 . 基于时空协同的多数据中心虚拟电厂低碳经济调度策略 [J]. 电力系统自动化 , 2024 , 48 ( 18 ): 56 - 65 .
Wen Xiyu , Zhu Jizhong , Li Shenglin , et al . Low-carbon economic dispatch for multiple data center virtual power plant based on spatio-temporal collaboration [J]. Automation of Electric Power Systems , 2024 , 48 ( 18 ): 56 - 65 .
Li C Y , Zheng K D , Guo H Y , et al . Computation-power coupled modeling for IDCs and collaborative optimization in ADNs [J]. IEEE Transactions on Smart Grid , 2024 , 15 ( 3 ): 2762 - 2775 .
张程 , 江炜龙 , 罗玉锦 , 等 . 考虑风光不确定性的多数据中心微网与共享储能协同优化运行 [J]. 电力系统保护与控制 , 2025 , 53 ( 4 ): 72 - 84 .
Zhang Cheng , Jiang Weilong , Luo Yujin , et al . Collaborative optimal operation of multiple data center microgrids and shared energy storage considering wind power and photovoltaic uncertainty [J]. Power System Protection and Control , 2025 , 53 ( 4 ): 72 - 84 .
王雨欣 , 陆海波 , 叶承晋 , 等 . 考虑多类型市场协同的互联数据中心运营策略 [J]. 电力系统自动化 , 2023 , 47 ( 24 ): 121 - 131 .
Wang Yuxin , Lu Haibo , Ye Chengjin , et al . Operation strategy for Internet data centers considering collaboration of multi-type markets [J]. Automation of Electric Power Systems , 2023 , 47 ( 24 ): 121 - 131 .
曹雨洁 , 丁肇豪 , 王鹏 , 等 . 能源互联网背景下数据中心与电力系统协同优化(二):机遇与挑战 [J]. 中国电机工程学报 , 2022 , 42 ( 10 ): 3512 - 3526 .
Cao Yujie , Ding Zhaohao , Wang Peng , et al . Coordinated operation for data center and power system in the context of energy Internet (Ⅱ): opportunities and challenges [J]. Proceedings of the CSEE , 2022 , 42 ( 10 ): 3512 - 3526 .
曹雨洁 . 数据中心和电网“算力-能源”协同优化及综合效益研究 [D]. 北京 : 华北电力大学 , 2024 .
Cao Yujie . Research on “computing power-energy” collaborative optimization and comprehensive benefits of data center and power grid [D]. Beijing : North China Electric Power University , 2024 .
Ding Z H , Chen S J , Sun Y M , et al . Data center job scheduling and energy management under uncertain environments [J]. IEEE Transactions on Industry Applications , 2025 , 61 ( 4 ): 5489 - 5500 .
Li J , Ma S Y , Xie M . A cluster-oriented task assignment optimization for green high-performance computing center operations [J]. Computers & Industrial Engineering , 2025 , 203 : 110929 .
Lima D H S , Aquino A L L , Curado M . Efficient task allocation in data center environments based on information theory [J]. Journal of Network and Systems Management , 2025 , 33 ( 3 ): 66 .
Ma J H , Yao R Y , Zhang B C , et al . Data-driven flexibility capability modeling of Internet data center considering task dependency [J]. IEEE Internet of Things Journal , 2024 , 11 ( 14 ): 24538 - 24550 .
Liu W Y , Yan Y J , Sun Y M , et al . Online job scheduling scheme for low-carbon data center operation: an information and energy nexus perspective [J]. Applied Energy , 2023 , 338 : 120918 .
Hu Y Y , Yang J , Ruan X L , et al . Green optimization for micro data centers: task scheduling for a combined energy consumption strategy [J]. Applied Energy , 2025 , 393 : 126031 .
Cao Y J , Cheng M , Zhang S F , et al . Data-driven flexibility assessment for Internet data center towards periodic batch workloads [J]. Applied Energy , 2022 , 324 : 119665 .
Guo J , Chang Z H , Wang S , et al . Who limits the resource efficiency of my datacenter: an analysis of Alibaba datacenter traces [C]// 2019 IEEE/ACM 27th International Symposium on Quality of Service (IWQoS) , 2019 : 1 - 10 .
Zheng W , Sakellariou R . Stochastic DAG scheduling using a Monte Carlo approach [J]. Journal of Parallel and Distributed Computing , 2013 , 73 ( 12 ): 1673 - 1689 .
Hussain M , Wei L F , Rehman A , et al . Deadline-constrained energy-aware workflow scheduling in geographically distributed cloud data centers [J]. Future Generation Computer Systems , 2022 , 132 : 211 - 222 .
Han M L , Jin X , Su X B , et al . Novel response-time bounds of typed DAG tasks on heterogeneous multicores [J]. IEEE Internet of Things Journal , 2025 , 12 ( 12 ): 19688 - 19701 .
Jypthi S A , Curino C , Menache I , et al . Morpheus: towards automated SLOs for enterprise clusters [C]// Proceedings of the 12th USENIX Conference on Operating Systems Design and Implementation , 2016 : 117 - 134 .
朱介北 , 刘东昊 , 李炳森 , 等 . 基于CPU工作状态进行温度估算的数据中心服务器功耗模型 [J]. 电力系统自动化 , 2023 , 47 ( 14 ): 140 - 148 .
Zhu Jiebei , Liu Donghao , Li Bingsen , et al . Data center server power consumption model based on temperature estimation considering CPU working state [J]. Automation of Electric Power Systems , 2023 , 47 ( 14 ): 140 - 148 .
Li W W , Qian T , Zhao W , et al . Decentralized optimization for integrated electricity-heat systems with data center based energy hub considering communication packet loss [J]. Applied Energy , 2023 , 350 : 121586 .
Liu Z M , Wang Y X , Vaidya S , et al . KAN: Kolmogorov-Arnold networks [PP/OL]. [ 2025-06-26 ]. https://arxiv.org/abs/2404.19756 https://arxiv.org/abs/2404.19756 .
Lin C H , Xu Y , Yi Z K , et al . Energy management for rural microgrid with inaccurate equipment parameters: a KAN-based deep reinforcement learning method [J]. IEEE Transactions on Smart Grid , 2025 , 16 ( 6 ): 4808 - 4821 .
Wang S X , Shi J R , Yang W , et al . High and low frequency wind power prediction based on Transformer and BiGRU-Attention [J]. Energy , 2024 , 288 : 129753 .
Cai Z C , Li X P , Ruiz R , et al . A delay-based dynamic scheduling algorithm for bag-of-task workflows with stochastic task execution times in clouds [J]. Future Generation Computer Systems , 2017 , 71 : 57 - 72 .
赵书强 , 赵蓬飞 , 韦子瑜 , 等 . 数据驱动下考虑多预测误差带信息的多场景随机优化调度 [J]. 电力自动化设备 , 2024 , 44 ( 11 ): 52 - 59 .
Zhao Shuqiang , Zhao Pengfei , Wei Ziyu , et al . Multi-scenario stochastic optimal scheduling considering multi-prediction error band information under data-driven [J]. Electric Power Automation Equipment , 2024 , 44 ( 11 ): 52 - 59 .
阿里巴巴 . 阿里巴巴集群跟踪计划 [EB/OL].[ 2025-05-21 ]. https://github.com/alibaba/clusterdata/tree/v2018/cluster-trace-v2018 https://github.com/alibaba/clusterdata/tree/v2018/cluster-trace-v2018 .
Alibaba . Alibaba cluster trace program [EB/OL]. [ 2025-05-21 ]. https://github.com/alibaba/clusterdata/tree/v2018/clust er-trace-v2018 https://github.com/alibaba/clusterdata/tree/v2018/cluster-trace-v2018 .
Tay Y , Dehghani M , Bahri D , et al . Efficient transformers: a survey [J]. ACM Computing Surveys , 2023 , 55 ( 6 ): 1 - 28 .
Xiang H R , Wu Z C , Wang R , et al . Robust support vector ordinal regression [J]. Information Sciences , 2025 , 717 : 122277 .
0
浏览量
0
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621