利用基因演算法對機房散熱進行配置最佳化

機構典藏 > College of Engineering > Graduate Institute & Department of Aerospace Engineering > Thesis > Item 987654321/94512

Please use this identifier to cite or link to this item: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/94512

Title:	利用基因演算法對機房散熱進行配置最佳化
Other Titles:	Reserch on spacing optimization for temperature reduction in a data center using genetic algorithms
Authors:	陳永祥;Chen, Yung-Hsiang
Contributors:	淡江大學航空太空工程學系碩士班湯敬民;Tang, Jing-Min
Keywords:	基因演算法;散熱;最佳化;Genetic Algorithms;Cooling;Optimization;FDS
Date:	2013
Issue Date:	2014-01-23 14:43:31 (UTC+8)
Abstract:	隨著全球能源枯竭導致能源成本快速增加，台灣多數能源無法自給自足，需仰賴國外進口；國際能源價格又節節攀升，國內民生物價也隨之水漲船高，造成人民怨聲不斷。近年來因為溫室效應、聖嬰現象等問題，使得環保意識抬頭、綠色節能減碳的議題逐漸受到各界的重視。因數據中心之維運需要大量的能源，如何有效利用使能源效率最大化、並降低其營運成本以達到節能減碳的目的為本研究的主要目標。本文主要利用基因演算法建置散熱最佳的機房散熱配置，節省機房的能源損耗，提高機房的散熱效率以達到節能省碳之效果。本研究利用機櫃間的走道間距、冷氣擺放位置及冷氣出風口的溫度做為參數進行機房內溫度差異的比較，首先利用走道做為最初染色體所得到的最佳結果，再與接下來的參數做交配，交配後比較其結果得到最佳配置方式。在本研究案例比較下得知，主要影響機房內溫度差異最大的因素為機房內流場的變化。由FDS觀察流場變化可見冷氣擺放位置影響機房溫度的差異，遠比改變走道間距差異大。本研究中之最佳的配置案例顯示，此例之機房內的對流循環良好、整體溫度分佈較其他案例平均，且此案例配置中冷氣直接對著溫度較高的熱通道進行吹送，可達到良好的降溫效果。 The energy is much more expensive nowadays because of the global energy depletion. Most of the energy needed in Taiwan has to be imported from the foreign country. With the global oil prices continue to rise, domestic oil prices is correspondingly getting higher. Due to the issues of greenhouse effect, El Nino, etc, people starts to concern about the subjects of eco-friendly and energy-saving. However, the data center consumes lots of energy while operating. Therefore, how to maximize the efficiency of consuming energy and lower the cost of operation of a data center is the main subject of this research. In this research, we used the genetic algorithms to build an optimized cooling configuration of a data center. The controlling parameter included were the spacing between the cabinets, the placement of air conditioner, and the temperature of the cold air flow. The optimized results obtained by previous simulation as well as the placement of air conditioners were used as controlling parameters of the successive computation to obtain an optimized configuration of the data center. The main cause of temperature variation in the room is found to be the moving flow in the data center. By flow patterns observations, the effect of temperature reduction caused by the relocation of air conditioner is higher than the spacing changes between aisles. Final results showed that the placement of the air conditioner was the most important parameters for temperature reduction, since the flow can be blown directly into the high temperature aisles and hence cooling the field.
Appears in Collections:	[Graduate Institute & Department of Aerospace Engineering] Thesis

Files in This Item:

File	Size	Format
index.html	0Kb	HTML	302	View/Open

Loading...