Data Center Thermal Environment Measurement Analysis And Optimization Transformation Based On Evaluation Indicators

Since twenty-first Century,with the rapid development of technology innovation driven Internet of things,AI,VR and other fields,data storage,cloud computing and network traffic have increased significantly.With the coming of big data era,data center construction is in full swing.At the same time,as an energy intensive facility,the development of servers in the data center from the chassis to the blade leads to a sharp increase in the heating density of a single cabinet.The data center has many heat dissipation devices,large heat dissipation,long equipment running time and strict requirements on the temperature of the equipment,which has caused huge energy consumption in the data center.The energy consumption of the data center air conditioning system accounts for about 40%of the total energy consumption of the data center,which is the main source of energy consumption in the data center.It is of great significance to build a green and efficient data center to save energy and reduce the consumption of the data center.By improving the thermal environment of the data center and adjusting the distribution of air flow,it can not only guarantee the needs of the safe operation of the server,but also reduce the energy consumption of the air conditioning system.To achieve this goal,domestic and foreign experts and scholars have conducted a lot of research and practice,the present research concentrates on the geometry optimization,improve the local hot spots,but also actively expand the evaluation index of thermal environment,in order to improve the thermal environment evaluation system,to better evaluate the data center airflow organization.The development of the data center has formed the frame "face to face,back to back"placed in the form,which has become the industry consensus.This arrangement can separate cold and hot passages and effectively prevent mixing of cold and hot gas flow.On this basis,the closed channel can better isolate the cold and hot gas flow.At present,the structure of cold channel closed-floor air supply is the most widely used in practical engineering.For the underfloor air supply data center,the air flow in the raised floor plays a key role in the effective operation of the data center.Therefore,the key to improve the thermal environment of the underfloor air supply data center is to optimize the distribution of the geometric structure in the static pressure layer to optimize the distribution of the air flow.Based on the measurement of the thermal environment of the existing cold channel closed floor to air supply data center,this paper studies the influence of geometric factors on airflow organization from the air inlet,floor perforation rate and static pressure baffle of the machine room.First of all,influence on room level factors--the air outlet is adjusted,the results showed that when the air inlet in the hot passage below,the perforated floor wind speed is bigger,the heat exchange between the cold air flow and the server is more sufficient,and the utilization of cold quantity is better.On this basis,the influence of the geometric structure under the elevated floor to the thermal environment of the data center is further studied.The perforation rate of the floor is measured and analyzed from the various structures of 10%?40%(10%change per time)and static pressure baffle angle from 30°?60°(15 0 per change).By comparing and analyzing the distribution of the temperature field and the velocity field,the combination of the best floor perforation rate and the baffle angle of the data center is selected.The results show that the distribution of the airflow in the data center is the most reasonable,the overall temperature distribution is uniform,and the local hot spot area and the scope are the smallest when the air inlet of the air conditioning is located below the thermal passage,the perforation rate of the floor is 30%,and the baffle angle is 60 degrees.While verifying the thermal environment in the measured data center velocity field and temperature field,this paper uses the thermal environment assessment index to evaluate the local thermal environment.Combined with the existing evaluation system,a new evaluation index LII(Local Integrated Index)is proposed to evaluate the local thermal environment.Through the evaluation index,we can further verify the results of optimization and transformation,and get the qualitative analysis conclusion of the influence of the geometry of the duct on the thermal environment of the cold-aisle containment data center underfloor air supply mode,so as to provide some references for the related engineering practice.