With the rapid development of new generation information technology such as5 G and metaverse,the annual power consumption of national data centers grows from125 billion k W·h to 270 billion k W·h from 2017 to 2022,of which the air conditioning energy consumption accounts for about 40% of the total energy consumption,and the unreasonable airflow organization is the main factor of excessive air conditioning energy consumption.This paper adopts a research method combining experimental test and simulation calculation by 6Sigma DC software.Through experimental test,the inlet air temperature distribution of racks at different locations,the uniformity of air supply through perforated floor and the distribution of temperature and humidity in the data room are explored,and relevant thermal environment indexes are borrowed for analysis;in the simulation analysis,the data room model and rack model are established respectively,and optimization analysis is carried out at different scales,and the evaluation of The thermal environment in the data room is evaluated by the indicators.Through the analysis of the measured data in the data room,it is found that the farther the rack position is from the air conditioner,the smaller and more uniform the inlet air temperature in the rack plane;compared with the rack at the edge position,the inlet air temperature in the rack at the center position is smaller and more uniform;there is no significant difference in the average inlet air temperature of the rack at the corresponding positions on both sides of the cold channel.With the perforated floor position farther away from the air conditioner the greater its air output,holes or gaps seriously affect the uniformity of the floor air supply,the air conditioning position has a certain impact on the uniformity of the floor air supply.The thermal environment indexes such as SHI,RTI and RCI of the data room were calculated,and it was found that there is hot air recirculation and cold air short circuit in the rack,and the thermal environment of the data room is poor.In the simulation of the data room,the results of temperature distribution and thermal environment indexes of the data room under nine cases of different air conditioning positions,closed single channel and closed both cold and hot channels are analyzed and compared.The results show that the closed channel can effectively prevent the mixing of hot and cold air and improve the velocity field and temperature field distribution in the data room.Compared with the closed single channel,the optimization effect is better when both cold and hot channels are closed,and the overall temperature decreases by 1℃~1.5℃ compared with the closed single;SHI,RHI,RTI and RCI are improved to different degrees when the channels are closed,and the RTI is reduced to 105% when the single channel is closed,and the optimization effect is better when both cold and hot channels are closed,and the RTI can reach 102%;the SHI value is higher when the single channel is closed compared with the cold and hot channels.Compared with the closed single channel and the closed cold and hot channels,the difference in SHI values is not significant,and the distribution is between 0.07 and 0.1;after the closed channels,the overall effect of the data room is best when running air conditioning 1-2 or air conditioning 2-3,and the second best when running air conditioning 1-3.In the rack simulation,the effects of different server installation lengths,perforated floor aperture rates and perforated floor air supply imbalance rates on the local airflow organization of the rack were compared and analyzed.The results show that the "all-short" and "top-length-bottom-short" IT equipment arrangements show the best heat dissipation and reduce the occurrence of local hot spots;When the floor aperture rate is ≤ 30%,the temperature and velocity fields of the air inlet plane of the rack are better;the increase of the unbalanced air supply rate will lead to the increase of the exhaust temperature of the IT equipment in the left and right side of the rack,but the impact on the side with reduced flow is greater;the IOM value of the computer rack shows that: the arrangement of the IT equipment with the "all-short" and "top-length-bottom-short" has the lowest probability of producing local hot spots.The probability of local hot spots is the smallest;with the increase of the floor opening rate and the unbalanced rate of perforated floor air supply,the IOM value of the rack is also increasing,and the probability of local hot spots increases. |