首页|算力-热力灵活性协同的数据中心能量管理方法

算力-热力灵活性协同的数据中心能量管理方法

扫码查看
随着数字经济产业快速发展,数据中心成为电能消耗的重要部分.多样化的数据处理需求、大量的计算用电及制冷用电,在带来高能耗问题的同时也为数据中心的运行调度提供了灵活性.因此,协同考虑计算灵活性和热灵活性,制定数据中心优化运行策略,可有效提高能源利用效率,提升数据中心运行经济性.该文提出一种考虑算力-热力灵活性协同的数据中心能量管理方法.首先,考虑数据负载灵活调度、冷却系统温度变化以及冷热通道热惯性等精细化要素,实现了信息技术设备和冷却设备的协同建模.其次,进行模型的线性化转换,降低了求解难度.在此基础上,实现了基于灵活性协同的数据中心能量管理,从而降低数据中心运行能耗及成本.最后通过算例验证了所提方法的有效性.
Optimal Energy Management of Data Centers Considering the Synergy of Computing Power and Thermal Power Flexibility
With the rapid development of the digital economy,data centers have become an important part of electricity consumption.The diversity of data processing demands and the large amount of electricity used by computing and cool-ing may result in high energy consumption problems and can provide flexibility for the scheduling of data center operations.Therefore,the operation strategy that considers both computing power and thermal power flexibility can ef-fectively improve energy efficiency and enhance the economy of data center operation.This paper proposes an optimal energy management strategy of data centers considering the synergy of computing power and thermal power flexibility.Firstly,the factors such as flexible load dispatch,cooling system temperature variations and thermal inertia in hot and cold aisles are taken into consideration,and collaborative modeling of information technology(IT)and cooling equipment is implemented.Secondly,the models are linearized to make it easy to solve.Finally,the energy management of data center based on the synergy of flexibility is realized,thereby reducing energy consumption and cost.The numerical results in case study validate the effectiveness of the proposed method.

data centerdata load dispatchthermal flexibilityrefined modelingenergy managementeconomic opera-tion

王天琪、于浩、赵金利、宋关羽、习伟、李鹏

展开 >

智能电网教育部重点实验室(天津大学),天津 300072

南方电网新型电力系统(北京)研究院有限公司,北京 102209

数据中心 数据负载调度 热灵活性 精细化建模 能量管理 经济运行

国家自然科学基金天津市自然科学基金

5230713222JCZDJC00700

2024

高电压技术
中国电力科学研究院 中国电机工程学会

高电压技术

CSTPCD北大核心
影响因子:2.32
ISSN:1003-6520
年,卷(期):2024.50(9)
  • 9