Multi-Objective Optimal Energy Saving Control for Air Conditioning System in Data Center

• Main Authors: Chun-Hao Huang, 黃軍浩
• Other Authors: Leehter Yao
• Format: Others
• Language: zh-TW
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/p38zqc

碩士 === 國立臺北科技大學 === 自動化科技研究所 === 105 === In recent years, the data center of the rapid development of the computer room, so need more space to store Information Technology (IT) equipment. As a result of the computer room air conditioners in cooling are large-scale in today’s, coupled with the server set with a high density, although the separation of hot and cold channels can effectively control the ambient temperature to the set range, but the actual IT equipment in the cooling effect is often limited so difficult to emit heat, continuing to reduce air-conditioning temperature often results in excessive energy consumption. So this paper discusses the cloud room will be developed for the air-conditioning system to meet the indoor temperature and energy saving multi-objective optimization strategy, First, the power consumption model and the temperature prediction model of the chiller system are established by using the Neural Network as the target function, Non-dominated sorting genetic algorithm (NSGA-II) to solve the chilled water supply temperature, fan operating rate of the best set point, In the case that satisfy indoor temperature, through the Power Usage Effectiveness(PUE) and the American Society of Heating, Refrigerating, and Air-Conditioning Engineers (ASHRAE) has recommended the RCI(Rack Cooling Index) as the best choice of the selection mechanism.