引用本文
  •    [点击复制]
  •    [点击复制]
【打印本页】 【下载PDF全文】 查看/发表评论下载PDF阅读器关闭

←前一篇|后一篇→

过刊浏览    高级检索

本文已被:浏览 148次   下载 234 本文二维码信息
码上扫一扫!
基于盲板的地板送风式数据中心热管理与能效案例研究
高启明1, 张羽2, 冯秀敏2, 张忠斌2
0
(1.合肥通用机械研究院有限公司;2.南京师范大学能源与机械工程学院)
摘要:
数据中心机房空调设备长期运行,每年需对其性能进行测试和评估,以确保安全高效运行。针对机架内服务器布局、盲板结构以及工况适应性进行实测,研究了封闭式冷通道地板下送风式数据中心的热环境;并通过热性能指标和能效指标,对该数据中心的安全性和能效进行了全年评估。结果表明:在冷通道中间区域推荐安装功率为2 kW以上的机架;机架间隙处安装盲板后,能够促进服务器内部冷气流循环,减少热气流的回流干扰,最大出口温度降低3.32 ℃,出口平均速度降低约24%;夏季工况时,该数据中心的PUE、WUE、CUE分别约为1.2、3.5、0.84,且WUE呈现出较强的季节性,冬季工况时,自由冷却可有效降低数据中心的能耗。此外,数据中心冷通道末端的机架过热问题显著,其中16个机架的机柜冷却指数均低于90%,热损失较高。
关键词:  数据中心  送风方式  机房空调  热性能评价指标  能源效率
DOI:
投稿时间:2023-05-04  修订日期:2023-06-27   录用日期:2023-08-09
基金项目:
Case Study on the Thermal Management and Energy Efficiency of Under-floor Air Distribution Data Centers Using Blind Plates
Gao Qiming1, Zhang Yu2, Feng Xiumin2, Zhang Zhongbin2
(1.Hefei General Machinery Research Institute Co., Ltd.;2.School of Energy and Mechanical Engineering, Nanjing Normal University)
Abstract:
Data center computer room air-conditioning equipment operates for prolonged periods, and its performance must be tested and evaluated annually to ensure safe and efficient operation. This study conducts field measurements on the server layout, blind plate structure, and working condition adaptability in the rack. The thermal environment of a data center using the closed cold aisle underfloor air distribution system is studied, and the thermal environment safety and energy efficiency throughout the year are evaluated through thermal performance indicators and energy efficiency indicators. The results show that it is recommended to install a rack with a power of more than 2 kW in the middle area of the cold aisle. Installing a blind plate in the gap between the racks can promote the circulation of cold air inside the server, reduce the backflow interference of hot air, and reduce the maximum outlet temperature by 3.32 °C. The average supply air speed was reduced by approximately 24%. Under summer operating conditions, the PUE, WUE, and CUE of the data center were approximately 1.2, 3.5, and 0.84, respectively, and the WUE exhibited strong seasonality. Under winter operating conditions, free cooling can effectively reduce the energy consumption of data centers. In addition, the overheating problem of the racks at the end of the cold aisle of the data center is significant. The cabinet cooling index of the 16 racks was less than 90%, and the heat loss was high.
Key words:  data center  air supply scheme  server room air conditioning  thermal performance evaluation index  energy efficiency

用微信扫一扫

用微信扫一扫