考虑负载响应特性的数据中心综合能源系统优化调度研究

doi:10.3969/j.issn.2097-0706.2023.10.004

摘要/Abstract

摘要：

为实现碳达峰、碳中和目标下的“算力+电力”协同，数据中心作为一种增长迅速、耗电量大的新型负载进行节能降碳，充分挖掘其调节潜力具有重大意义。为充分发挥数据中心负载灵活性，提出一种考虑负载响应特性的数据中心综合能源系统优化调度的方法。首先，建立了数据中心综合能源系统优化调度框架，对数据中心的负载响应特性和能耗设备进行建模，对数据中心冷、热、电3种负荷展开研究；其次，确定目标函数和约束条件，以数据中心能源系统运行成本最低为目标建立优化模型；最后，以某数据中心为例进行仿真分析，通过对算例进行不同场景对比分析发现，最佳策略能够有效降低系统成本和能耗，实现工作负载的最佳分配。

关键词: 碳中和, 数据中心, 负载响应, 优化调度, 综合能源, 节能降碳

Abstract:

In the collaboration of "computing power + electric power" under the goal of carbon peaking and carbon neutrality，the data center，whose power consumption is substantive and growing rapidly，has great potential in carbon emission reduction and load regulation. To give full play to the flexibility of data center loads，an optimal scheduling strategy of the data center integrated energy system considering the load response characteristics is proposed. Firstly，an optimized scheduling framework for the data center integrated energy system is established. The load response characteristics and equipment energy consumption of the data center are modelled，to study the cold，heat and electric loads of the data center. Then，the objective function and constraint conditions are determined，and the optimization model is established to minimize the operation cost of the data center energy system. Finally，the simulation analysis is made on a data center. Comparing the performances of the data center energy system under different scenarios， the optimal strategy with the optimal allocation of workloads is figured out which can effectively reduce the system cost and energy consumption.

Key words: carbon neutral, data center, load response, optimized scheduling, integrated energy, energy saving and carbon reduction

中图分类号:

TK01

余文昶, 陈永刚, 曹俊波, 左鹿原, 张相寅, 杨秀. 考虑负载响应特性的数据中心综合能源系统优化调度研究[J]. 综合智慧能源, 2023, 45(10): 25-34.

YU Wenchang, CHEN Yonggang, CAO Junbo, ZUO Luyuan, ZHANG Xiangyin, YANG Xiu. Optimal scheduling of the data center integrated energy system considering load response characteristics[J]. Integrated Intelligent Energy, 2023, 45(10): 25-34.

图/表 14

图1

表1

服务器集群特性

数量	CPU型号	f/Hz	服务率	$P a d d$	$K$	最大延迟时间/h	运维/投资成本/（元·台^-1）
1 500	AMD速龙	2	13.5	53	6	10	150/15 000

表1

表2

图2

图3

图4

表3

CARS相关参数

名称	参数	值
送风机	$a 1 / b 1 / c 1 / d 1 / e 1 / α f a n / ρ a i r / q m, a_N$	0.35/0.31/-0.54/0.87/0.00/0.70/1.29 kg/m³/52.55 kg/s
冷冻水泵	$a 2 / b 2 / c 2 / d 2 / q m, e_N / P e_N / c p$	0/1/0/0/31.00 kg/s/7.93 kW/4.19 kJ/(kg⋅℃)
制冷机	$a 3 / b 3 / c 3 / d 3 / e 3 / f 3 / a 4 / b 4 / c 4 / d 4 / e 4 / f 4 / a 5 / b 5 / c 5 / Q N / η C O P, N$	0.44/-0.081/-0.002/0.092/-0.0031/0.007/0.05/-0.016/-0.000 3/0.024/-0.000 3/-0.000 6/0.188/0.356/0.454/720 kW/6.04
冷却水泵	$P c_N$	9.17 kW
冷却塔	$P c t_N$	8.82 kW
机房室内送风温度	$t o u t, m i n / t o u t, m a x$	23 ℃/23 ℃

表3

图5

图6

图7

图8

图9

图10

表4

参考文献 24

[1]	郭创新, 王惠如, 张伊宁, 等. 面向区域能源互联网的“源-网-荷”协同规划综述[J]. 电网技术, 2019, 43(9):3071-3080.
	GUO Chuangxin, WANG Huiru, ZHANG Yining, et al. Review of "source-net-load" collaborative planning for regional energy internet[J]. Power Grid Technology, 2019, 43(9):3071-3080.
[2]	曾博, 白婧萌, 郭万祝, 等. 智能配电网需求响应效益综合评价[J]. 电网技术, 2017, 41(5):1603-1611.
	ZENG Bo, BAI Jingmeng, GUO Wanzhu, et al. Comprehensive evaluation of demand response benefits of intelligent distribution network[J]. Power Grid Technology, 2017, 41(5): 1603-1611.
[3]	赵福林, 俞啸玲, 杜诗嘉, 等. 计及需求响应的含大规模风电并网下电力系统灵活性评估[J]. 电力系统保护与控制, 2021, 49(1):42-51.
	ZHAO Fulin, YU Xiaoling, DU Shijia, et al. An assessment of the flexibility of off-grid power systems with large-scale wind power in consideration of demand response[J]. Power System Protection and Control, 2021. 49(1): 42-51.
[4]	罗金满, 赵善龙, 封祐钧, 等. 考虑综合需求响应不确定性的电-气综合能源系统优化运行[J]. 中国电力, 2020, 53(12):119-126.
	LUO Jinman, ZHAO Shanlong, FENG Youjun, et al. Optimal operation of an integrated power-gas energy system considering the uncertainty of comprehensive demand response[J]. China Electric Power, 2020, 53(12): 119-126.
[5]	DING Z, CAO Y, XIE L, et al. Integrated stochastic energy management for data center microgrid considering waste heat recovery[J]. IEEE Transactions on Industry Applications, 2019, 55(3): 2198-2207. doi: 10.1109/TIA.28
[6]	WANG P, CAO Y, DING Z. Flexible multi-energy scheduling scheme for data center to facilitate wind power integration[J]. IEEE Access, 2020, 8: 88876-88891. doi: 10.1109/Access.6287639
[7]	陈敏, 高赐威, 陈宋宋, 等. 考虑数据中心用电负荷调节潜力的双层经济调度模型[J]. 中国电机工程学报, 2019, 39(5):1301-1313.
	CHEN Min, GAO Ciwei, CHEN Songsong, et al. A two-layer economic scheduling model considering the potential of power load regulation in data centers[J]. Proceedings of the CSEE, 2019. 39(5): 1301-1313.
[8]	YU L, JIANG T, CAO Y, et al. Carbon-aware energy cost minimization for distributed internet data centers in smart microgrids[J]. IEEE Internet of Things Journal, 2014, 1(3):255-264. doi: 10.1109/JIOT.2014.2322606
[9]	YU Y, JIANG T, ZOU Y. Real-time energy management for cloud data centers in smart microgrids[J]. IEEE Access, 2017, 4:941-950. doi: 10.1109/ACCESS.2016.2539369
[10]	YU L, JIANG T, ZOU Y L. Distributed real-time energy management in data center microgrids[J]. IEEE Transactions on Smart Grid, 2018, 9(4):3748-3762. doi: 10.1109/TSG.2016.2640453
[11]	HOGADE N S, PASRICHA S, SIEGEL H J, et al. Minimizing energy costs for geographically distributed heterogeneous data centers[J]. IEEE Transactions on Sustainable Computing, 2018:318-331.
[12]	雍培, 张宁, 慈松, 等. 5G通信基站参与需求响应:关键技术与前景展望[J]. 中国电机工程学报, 2021, 41(16):5540-5551.
	YONG Pei, ZHANG Ning, CI Song, et al. Participation of 5G communication base stations in demand response:Key technologies and prospects[J]. Proceedings of the CSEE, 2021. 41(16): 5540-5551.
[13]	兰洲, 蒋晨威, 谷纪亭, 等. 促进可再生能源发电消纳和碳减排的数据中心优化调度与需求响应策略[J]. 电力建设, 2022, 43(4):1-9. doi: 10.12204/j.issn.1000-7229.2022.04.001
	LAN Zhou, JIANG Chenwei, GU Jiting, et al. Data center optimal scheduling and demand response strategy for promoting renewable energy power consumption and carbon emission reduction[J]. Electric Power Construction, 202, 43(4):1-9.
[14]	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. Research on carbon neutrality regulation method of interconnected multi-data centers based on time-space dual-dimensional migration of computing load[J]. Proceedings of the CSEE, 202, 42(1):164-177.
[15]	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. A dual-layer optimization model for multiple data centers considering load characteristics and response characteristics[J]. Automation of Electric Power Systems, 2022, 46(21):30-41.
[16]	曾博, 刘一贤, 张卫翔, 等. 综合考虑内外生不确定性的绿色数据中心多域资源协同规划方法[J/OL]. 中国电机工程学报:1-18(2023-01-31)[2023-05-16]. http://kns.cnki.net/kcms/detail/11.2107.TM.20230130.1641.003.html.
	ZENG Bo, LIU Yixian, ZHANG Weixiang, et al. Multi-domain resource collaborative planning method for green data centers considering internal and external uncertainties[J/OL]. Proceedings of the Chinese Society for Electrical Engineering: 1-18(2023-01-31)[2023-05-16]. http://kns.cnki.net/kcms/detail/11.2107.TM.20230130.1641.003.html.
[17]	周吟雨, 董厚琦, 曾博, 等. 考虑灵活性潜力的互联网数据中心与配电网双层协同规划方法[J]. 电力系统保护与控制, 2022, 50(24):49-59.
	ZHOU Yinyu, DONG Houqi, ZENG Bo, et al. A two-layer collaborative planning method for Internet data centers and distribution networks considering flexibility potential[J]. Power System Protection and Control, 2022, 50(24): 49-59.
[18]	高赐威, 吴刚, 陈宋宋. 考虑地理分散的数据中心服务器频率调节的电网降损模型[J]. 中国电机工程学报, 2019, 39(6):1673-1681.
	GAO Ciwei, WU Gang, CHEN Songsong. Network loss reduction model considering frequency regulation of geographically dispersed data center servers[J]. Proceedings of the CSEE, 2019, 39(6):1673-1681.
[19]	李志文, 李卫国, 王利利. 数据中心能耗分析研究[J]. 现代盐化工, 2018, 45(4):36-37.
	LI Zhiwen, LI Weiguo, WANG Lili. Energy consumption analysis of data centers[J]. Modern Salt Chemical Industry, 2018, 45(4): 36-37.
[20]	张志亮, 张斌宇. 北方某IDC数据机房空调冷水机组模型建立及能耗分析[J]. 长江信息通信, 2022, 35(10):183-186.
	ZHANG Zhiliang, ZHANG Binyu. Model establishment and energy consumption analysis of air conditioner chiller in an IDC data room in north China[J]. Yangtze River Information and Communication, 2022, 35(10):183-186.
[21]	耿圣杰, 贾燕冰, 江坷滕, 等. 电网数据中心服务器容量及其综合供能系统联合规划策略研究[J]. 电网技术, 2022, 46(9):3281-3292.
	GENG Shengjie, JIA Yanbing, JIANG Keteng, et al. Research on joint planning strategy of power grid data center server capacity and integrated energy supply system[J]. Power Grid Technology, 2022, 46(9):3281-3292.
[22]	祁兵, 曹望璋, 李彬, 等. 考虑托管式数据中心负荷调节不确定性的区间优化模型[J]. 电网技术, 2022, 46(1):39-49.
	QI Bing, CAO Wangzhang, LI Bin, et al. Interval optimization model considering load regulation uncertainty of managed data center[J]. Power Grid Technology, 2022, 46(1):39-49.
[23]	吴云芸, 方家琨, 艾小猛, 等. 计及需求响应的数据中心联盟共享储能规划[J]. 电力系统自动化, 2023, 47(7):42-50.
	WU Yunyun, FANG Jiakun, AI Xiaomeng, et al. Shared energy storage planning for data center alliances considering demand response[J]. Automation of Electric Power Systems, 2023, 47(7):42-50.
[24]	付豪, 邹花蕾, 张腾飞. 基于长短期记忆循环神经网络的变电站监控系统智能故障推理方法[J]. 综合智慧能源, 2022, 44(12): 11-17. doi: 10.3969/j.issn.2097-0706.2022.12.002
	FU Hao, ZOU Hualei, ZHANG Tengfei. Intelligent fault reasoning method for the substation monitoring system based on LSTM[J]. Integrated Intelligent Energy, 2022, 44(12): 11-17. doi: 10.3969/j.issn.2097-0706.2022.12.002

设备名称	参数	值	投资/运维成本/[元·（MW·h）^-1]
燃气轮机	功率上/下限、产电效率、天然气低位热值	1 500 MW/0 MW，0.184，8 000 kJ/m³	3 000/53
风机	出力上/下限	2 000 MW/0 MW	5 000/29
光伏	出力上/下限	2 200 MW/0 MW	4 000/25
电制冷机	出力上/下限、制冷系数	600 MW/0 MW，3.00	1 000/11
吸收式制冷机	出力上/下限、制冷系数	600 MW/0 MW，1.38	1 100/13
电储能	额定功率、出力上/下限、SOC初始值/上/下限	1 200 MW，500 MW/0.4 MW，0.2/0.2/0.9	960/30
热储能	额定功率、出力上/下限、SOC初始值/上/下限	1 200 MW，500 MW/0.4 MW，0.2/0.2/0.9	880/24
蓄冷槽	额定功率、出力上/下限、SOC初始值/上/下限	1 200 MW，500 MW/0.4 MW，0.2/0.2/0.9	860/26
主电网交互	功率上/下限	4 500 MW/0 MW	—

场景	购能成本	运维成本	投资成本	运行总成本
1	3 202.056 7	94.527 4	330.833 7	3 627.417 8
2	1 350.691 3	94.525 8	330.833 7	1 644.851 2
3	2 949.076 9	94.741 8	332.201 7	3 376.020 4
4	885.444 1	94.521 4	331.517 7	1 390.269 3