Research On The Application Of Reversible Fan Unit And Energy Optimization Analysis In The Cooling System Of Data Center

In recent years,the data center has a trend of high power density and highly dynamic,and the cooling system of data center is facing a serious thermal challenge.There is a common phenomenon that one or several server racks have a higher or lower heat power,and there is a mismatching between the volume flow rate of tile-supplied cooling air and the cooling requirement of server racks.In this thesis,the reversible fan unit was proposed to realize the individuation management of cooling resources in data center,and solve the problem of mismatching between supply and demand.The characteristics of cooling system(airflow organization,rack inlet temperature,power consumption,etc.)were investigated.The follow research works were carried out in this thesis:Firstly,the simulation model of the reversible fan unit was established and brought into the data center model.The impacts of H(the length of the duct)on the tiles flow rates were explored corresponding to each work pattern(supply upwards and working reversely)of the reversible fan system.The results indicated that when the length of the duct was 0.7 m,the fan unit can efficiently realize the individuation management of the local cooling resource,and the uniformity of the tile flow rates was improved.Secondly,the airflow organization and rack inlet temperature were investigated corresponding to each work pattern.For the situation that one or several server racks have a higher or lower heat power and the mismatching between supply and demand was existent,the reversible fan unit was applied to manage the local airflow and improve the airflow organization.In this part,the research work was carried out in two aspects(local region and whole cold aisle),and the impacts of the derating factor and fan-to-tile distance on the boundary/inner flow of the cold aisle and rack inlet temperature were investigated.Besides,the SHI of the 142 scenarios involved in this paper were computed,compared and analyzed.The results indicated that the airflow organization and rack inlet temperature can be improved by applying the reversible fan unit properly.Thirdly,optimization analysis of the reversible fan unit and the cooling system based on the Response Surface Methodology.In this part,the impacts of derating factor and fan-to-tile distance on the maximum rack inlet temperature corresponding to each work pattern of reversible fan unit were investigated by means of the Response Surface Methodology.The prediction models of the maximum rack inlet temperature were established and examined.Besides,the optimal working states of the reversible fan unit were figured out by analyzing the prediction models and the optimal working states were examined by the means of numerical solution.Finally,the energy efficiency of the cooling system was investigated,and the results indicated that the power consumption of the cooling system can be reduced 9.48%and 7.63%respectively corresponding to the two work pattern of the reversible fan unit.The power consumption of the cooling system in data center is vast,and the reversible fan unit investigated in this paper has a vital significance to realize the individuation management of the cooing resources and the energy saving of the cooling system in data center.