[1]李庆华,冉泳屹,刘启晨,等.数据中心冷热电联产系统的前摄式智能节能优化算法[J].智能系统学报,2025,20(1):139-149.[doi:10.11992/tis.202312037]
LI Qinghua,RAN Yongyi,LIU Qichen,et al.Proactive intelligent energy-saving optimization algorithm for data center CCHP system[J].CAAI Transactions on Intelligent Systems,2025,20(1):139-149.[doi:10.11992/tis.202312037]
点击复制
LI Qinghua,RAN Yongyi,LIU Qichen,et al.Proactive intelligent energy-saving optimization algorithm for data center CCHP system[J].CAAI Transactions on Intelligent Systems,2025,20(1):139-149.[doi:10.11992/tis.202312037]
数据中心冷热电联产系统的前摄式智能节能优化算法
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
20
期数:
2025年第1期
页码:
139-149
栏目:
学术论文—智能系统
出版日期:
2025-01-05
- Title:
- Proactive intelligent energy-saving optimization algorithm for data center CCHP system
- Keywords:
- green data center; combined cooling; heating and power generation; intelligent energy-saving; deep reinforcement learning; carbon emission optimization; energy efficiency improvement; joint control; predictive network
- 分类号:
- TP272
- 摘要:
- 现有的数据中心节能降碳优化方法没有综合考虑碳足迹涉及的能源输入、生产耗能以及废余利用等环节的耦合性，难以实现系统性节能降碳。为此，提出了一种基于深度强化学习的优化算法DeepCCHP(deep combined cooling, heating and power generation)，针对数据中心冷热电联产系统，联合控制供电子系统和制冷子系统，优化用电成本、碳排放量和能效。DeepCCHP结合长、短期时间序列网络和深度强化学习方法对联合优化问题进行求解，实现前摄式的联合控制发电设备和制冷设备。在基于Trnsys软件的仿真环境中，通过阿里巴巴数据中心集群数据的训练和验证。实验结果表明，与基准算法相比，DeepCCHP算法可以节省最高40%的成本和28%的碳排放量，且能够在能源成本、碳排放和能效三者之间取得更好的折中与平衡。
- Abstract:
- The existing methods for energy-saving and carbon reduction optimization in data centers lack a comprehensive consideration of the coupling of carbon footprint-related factors, including energy input, production consumption, and waste utilization. This limitation hinders the achievement of systematic energy-saving and carbon reduction. To address this issue, a deep reinforcement learning-based optimization algorithm, named DeepCCHP, is proposed. This algorithm focuses on the combined cooling, heating and power generation (CCHP) in data centers, employing coordinated control of the power supply and cooling systems to optimize electricity cost, carbon emissions, and PUE. DeepCCHP integrates a long and short-term time-series network-attention (LSTNet-Attn) for multi-dimensional time series forecasting and a deep reinforcement learning approach to solve the joint optimization problem, achieving proactive joint control of power generation and cooling equipment. The algorithm is validated through training and verification using Alibaba data center cluster data in a simulation environment based on Trnsys software. Experimental results demonstrate that, compared with baseline algorithms, the DeepCCHP algorithm can achieve up to 40% cost savings and 28% reduction in carbon emissions. It also demonstrates a better trade-off and balance among energy cost, carbon emissions, and energy efficiency.
- 参考文献/References:
-
[1] 李德毅. AI: 人类社会发展的加速器[J]. 智能系统学报, 2017, 12(5): 583-589.
LI Deyi. Artificial intelligence: an accelerator for the development of human society[J]. CAAI transactions on intelligent systems, 2017, 12(5): 583-589.
[2] 中国信息通信研究院: 数据中心白皮书(2022)[EB/OL]. (2022-04-22)[2023-12-24]. http://www.caict.ac.cn/kxyj/qwfb/bps/202204/t20220422_400391.html.
China Academy of Information and Communications Technology: data center white paper(2022)[EB/OL]. (2022-04-22)[2023-12-24]. http://www.caict.ac.cn/kxyj/qwfb/bps/202204/t20220422_400391.htm.
[3] 工业和信息化部电子第五研究所计量检测中心. 中国数字基建的脱碳之路: 数据中心与5G减碳潜力与挑战(2020—2035)[R]. 广州: 工业和信息化部电子第五研究所计量检测中心, 2021.
Metrology and Testing Center of the Fifth Research Institute of Electronics of the Ministry of Industry and Information Technology. China’s digital infrastructure decarbonization path: data centers and 5G carbon reduction potentials and challenges(2020—2035)[R]. Guangzhou: Metrology and Testing Center of the Fifth Research Institute of Electronics of the Ministry of Industry and Information Technology, 2021.
[4] 娄洁良. 数据中心能效案例及节能运行[J]. 电信科学, 2019, 35(2): 95-104.
LOU Jieliang. Data center energy efficiency case and energy saving operation[J]. Telecommunications science, 2019, 35(2): 95-104.
[5] 张沈习, 王丹阳, 程浩忠, 等. 双碳目标下低碳综合能源系统规划关键技术及挑战[J]. 电力系统自动化, 2022, 46(8): 189-207.
ZHANG Shenxi, WANG Danyang, CHENG Haozhong, et al. Key technologies and challenges of low-carbon integrated energy system planning for carbon emission peak and carbon neutrality[J]. Automation of electric power systems, 2022, 46(8): 189-207.
[6] 陈信甫, 梁珍. 风冷数据中心热回风预热市政冷水的节能与可行性分析[J]. 制冷与空调, 2023, 23(11): 80-88.
CHEN Xinfu, LIANG Zhen. Energy saving and feasibility analysis of municipal cold water preheating by hot return air in air-cooled data center[J]. Refrigeration and air-conditioning, 2023, 23(11): 80-88.
[7] 郑瑛. 云计算数据中心节能调度算法改进研究[J]. 西南大学学报(自然科学版), 2019, 41(12): 135-142.
ZHENG Ying. Improvement of energy-saving scheduling algorithm in cloud computing data centers[J]. Journal of Southwest University (natural science edition), 2019, 41(12): 135-142.
[8] 卜昆. “碳中和” 驱动下能源大数据中心建设的问题与对策[J]. 信息系统工程, 2023(12): 56-59.
BU Kun. Problems and countermeasures of energy big data center construction driven by “carbon neutralization”[J]. China CIO news, 2023(12): 56-59.
[9] WAN Jianxiong, ZHOU Jie, GUI Xiang. Sustainability analysis of green data centers with CCHP and waste heat reuse systems[J]. IEEE transactions on sustainable computing, 2021, 6(1): 155-167.
[10] WANG Ruihang, CAO Zhiwei, ZHOU Xin, et al. Green data center cooling control via physics-guided safe reinforcement learning[J]. ACM transactions on cyber-physical systems, 2024, 8(2): 1-26.
[11] HE Wei, ZHANG Jifang, LI Hailong, et al. Optimal thermal management of server cooling system based cooling tower under different ambient temperatures[J]. Applied thermal engineering, 2022, 207: 118176.
[12] YADAV R, ZHANG Weizhe, LI Keqin, et al. Managing overloaded hosts for energy-efficiency in cloud data centers[J]. Cluster computing, 2021, 24(3): 2001-2015.
[13] ZHOU Zhou, SHOJAFAR M, ALAZAB M, et al. AFED-EF: an energy-efficient VM allocation algorithm for IoT applications in a cloud data center[J]. IEEE transactions on green communications and networking, 2021, 5(2): 658-669.
[14] VENKATASWAMY V, GRIGSBY J, GRIMSHAW A, et al. Rare: renewable energy aware resource management in Datacenters[M]//Lecture Notes in Computer Science. Cham: Springer Nature Switzerland, 2023: 108-130.
[15] NGOBENI A, DANIEL CHOWDHURY S P. Electrical design for a combined cooling, heating and power for go-green data centers[C]//2018 IEEE PES/IAS Power Africa. Cape Town: IEEE, 2018: 901-906.
[16] LYU Jiawei, ZHANG Shenxi, CHENG Haozhong, et al. Optimal sizing of energy station in the multienergy system integrated with data center[J]. IEEE transactions on industry applications, 2021, 57(2): 1222-1234.
[17] ZHANG Tingjun, YUE Dong, YU Liang, et al. Joint energy and workload scheduling for fog-assisted multimicrogrid systems: a deep reinforcement learning approach[J]. IEEE systems journal, 2023, 17(1): 164-175.
[18] PRAVEENCHANDAR J, TAMILARASI A. RETRACTED ARTICLE: dynamic resource allocation with optimized task scheduling and improved power management in cloud computing[J]. Journal of ambient intelligence and humanized computing, 2021, 12(3): 4147-4159.
[19] RAN Yongyi, HU Han, WEN Yonggang, et al. Optimizing energy efficiency for data center via parameterized deep reinforcement learning[J]. IEEE transactions on services computing, 2023, 16(2): 1310-1323.
[20] CHI Ce, JI Kaixuan, SONG Penglei, et al. Cooperatively improving data center energy efficiency based on multi-agent deep reinforcement learning[J]. Energies, 2021, 14(8): 2071.
[21] RAN Yongyi, ZHOU Xin, HU Han, et al. Optimizing data center energy efficiency via event-driven deep reinforcement learning[J]. IEEE transactions on services computing, 2023, 16(2): 1296-1309.
[22] LI Longxi, YU Shiwei, MU Hailin, et al. Optimization and evaluation of CCHP systems considering incentive policies under different operation strategies[J]. Energy, 2018, 162: 825-840.
[23] GUPTA R, PURI I K. Waste heat recovery in a data center with an adsorption chiller: technical and economic analysis[J]. Energy conversion and management, 2021, 245: 114576.
[24] ZHANG Qingxia, MENG Zihao, HONG Xianwen, et al. A survey on data center cooling systems: technology, power consumption modeling and control strategy optimization[J]. Journal of systems architecture, 2021, 119: 102253.
[25] LILLICRAP T P, HUNT J J, PRITZEL A, et al. Continuous control with deep reinforcement learning[EB/OL]. (2015-09-09)[2023-12-24]. https://arxiv.org/abs/1509.02971v6.
[26] LAI Guokun, CHANG Weicheng, YANG Yiming, et al. Modeling long- and short-term temporal patterns with deep neural networks[C]//The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval. Ann Arbor: ACM, 2018: 95-104.
[27] 申翔翔, 侯新文, 尹传环. 深度强化学习中状态注意力机制的研究[J]. 智能系统学报, 2020, 15(2): 317-322.
SHEN Xiangxiang, HOU Xinwen, YIN Chuanhuan. State attention in deep reinforcement learning[J]. CAAI transactions on intelligent systems, 2020, 15(2): 317-322.
[28] WANG Dongxiao, XIE Changhong, WU Runji, et al. Optimal energy scheduling for data center with energy nets including CCHP and demand response[J]. IEEE access, 2021, 9: 6137-6151.
[29] MNIH V, KAVUKCUOGLU K, SILVER D, et al. Human-level control through deep reinforcement learning[J]. Nature, 2015, 518(7540): 529-533.
[30] XIANG Shilin, XIANG Yue, LU Yu, et al. Modeling and optimization of data center energy consumption[C]//2023 Panda Forum on Power and Energy. Chengdu: IEEE, 2023: 544-549.
备注/Memo
收稿日期:2023-12-24。
基金项目:国家自然科学基金项目（U23A20275, 62101525, 62171072）；重庆市自然科学基金项目（cstc2021jcyj-msxmX0586）.
作者简介:李庆华，硕士，主要研究方向为绿色数据中心、智能节能、深度强化学习。E-mail：s210131103@stu.cqupt.edu.cn。;冉泳屹，文峰副教授，博士，文峰青年百人，IEEE会员，主要研究方向为计算机及网络系统的智能控制与优化。主持国家自然科学基金区域联合基金重点项目课题、国家自然科学基金青年基金项目、重庆市自然科学基金面上项目、重庆市留创计划等科研项目4项，入选2021年重庆市留学人员回国创业创新支持计划。发表学术论文40余篇，发表英文专著章节1章，译著1部，获得3次国际会议最佳论文奖。E-mail：ranyy@cqupt.edu.cn。;刘启晨，硕士，主要研究方向为进阶建筑设计、建筑与景观。E-mail：1243163016@qq.com。
通讯作者:冉泳屹. E-mail：ranyy@cqupt.edu.cn
更新日期/Last Update:
2025-01-05