| Abstract: | 大數據時代的來臨，強悍的電腦運算速度、大量的資料儲存空間及高速的網際網路需求造就大量資料中心的興起。而在全球暖化、資源短缺的21世紀，如何將資料中心的耗能降到最低，即將資料中心的能源使用效率(PUE)，降到最低是現今非常重要的節能減排課題。
空調是資料中心耗能最大的來源，自然冷卻是降低空調耗能最佳的手段，本研究嘗試以熱管散熱器來達到資料中心自然冷卻的目的。本研究分成三部分，第一部分是以田口法做虹吸熱管的各種參數，包括真空度、填充率、展弦比、傾斜角做最佳化模擬。第二部分是將熱管組成熱管散熱器之參數，包括鰭片間距、鰭片厚度、熱管排列做最佳化模擬分析，第三部份是將熱管散熱器放置於資料中心內的參數，包括熱管散熱器的數量、熱管散熱器的配置做最佳化模擬分析。
最後的結果顯示本研究所採用之數值離散方法確實對於虹吸熱管物理機制及傳熱性能分析具有適用性，也與實驗值相當一致、經數值模擬後熱管散熱器性能與實驗值相當吻合及資料中心節能率實驗值也與數值模擬值非常接近，並能得到最佳熱管散熱器的配置與數量。本研究能以最佳化的熱管、熱管散熱器設計及熱管散熱器配置來實現最低的資料中心PUE及最佳的節能率。
Big data age, the advent of powerful computer computing speed, a lot of data storage space and high-speed Internet needs to create a large number of data center rise. In the 21st century, the global warming and resource shortages, how to minimize the energy consumption of the data center, that is, the data center PUE (Power Usage Effectiveness) to a minimum is now very important energy-saving emission reduction issues.
Air conditioning is the most power consumption of data center, Free Cooling is the best way to reduce the power consumption of air conditioning, this study attempts to heat pipe heat exchanger to achieve the purpose of Free Cooling data center. This study is divided into three parts, the first part is the Taguchi method to do various parameters of the thermosyphon, including the vacuum, filling rate, aspect ratio, incline angle to do the optimization simulation, the second part is the heat pipe heat exchanger parameters , including fin spacing, fin thickness, heat pipe arrangement to do the best simulation analysis, the third part is the heat pipe heat exchanger placed in the data center parameters, including the number of heat pipe heat exchanger, heat pipe heat exchanger configuration to do optimization simulation analysis. This study aims to achieve the lowest data center PUE (Power Usage Effectiveness) with optimized heat pipe heat exchanger arrangement and heat pipe heat exchanger design. |
