| With the advent of the 5G era,the market scale of data centers in my country has shown a trend of rapid growth,followed by the problems of high energy consumption and high thermal power density.Optimizing the airflow organization in the data center provides a new solution to effectively deal with the increasingly prominent problems.The optimization of heat dissipation in the data center is conducive to improving the cooling efficiency of computer room air conditioning units(CRACs).In addition,the increasing energy consumption of CRACs has also increased the load pressure on the local power grid.Therefore,it is particularly important to adopt demand-side management methods to control the CRACs.In this regard,combined with the geometric modeling of the data center,this paper conducts research from two aspects:the optimization for the cooling of the computer room space layout and the judgment of the interruption potential of the air-conditioning load.A demand response interaction method based on the cooling optimization of the data center is proposed.The specific research content of this paper is as follows.Firstly,the basic theory of CFD numerical simulation and the software Fluent for dealing with airflow organization and thermal environment are introduced.Based on the model assumption and model simplification of the data center,an improved rack-server model which is different from the black box model and can reflect the internal details of the rack is proposed.By comparing with the experimental data,the reliability and accuracy of the model in simulating the airflow temperature field and velocity field are verified.Secondly,based on the improved cabinet-server model,the air velocity field and temperature field in the computer room were qualitatively compared and analyzed under different air supply methods,whether considering cold aisle containment(CAC)methods,and different outlet positions.It is concluded that the best air flow organization is in the form of underfloor air supply,considering CAC,and outlet above the hot aisle.Based on the above arrangement and combination,the velocity and temperature non-uniformity coefficients for evaluating the air supply uniformity are introduced.The effects of CRACs layout,tile porosity and raised floor height on the supply air uniformity of the data center were quantitatively compared and analyzed respectively.It is concluded that the supply air uniformity relatively preferred under CRACs layout ΙΙΙ,25%-45% tile porosity and 0.55m-0.75 m raised floor height.Then,based on the orthogonal experimental design method and the range analysis method,the evaluation indicators are the non-uniformity coefficient of the tile supply air velocity,the rack inlet velocity,the rack inlet temperature and the blade server temperature.The factor-level combination optimization design is carried out for the CRACs layout,the tile porosity,and the raised floor height.The order of importance for each factor is obtained as the CRACs layout,the raised floor height and the tile porosity.CRACs layout ΙΙΙ,36% tile porosity and 0.75 m raised floor height is the optimal factor-level combination of the data center.The model basis and theoretical reference guide are provided for the construction of new data centers and the reconstruction of old data centers.Finally,based on the optimal factor-level combination operating conditions of the data center,combined with the hotspot temperature field of the blade server model,the maximum interruptible potential of the CRACs load is quantitatively analyzed from the following four aspects: different steady-state supply air temperatures,Different pre-cooling time,different supply air temperature during temperature recovery and different outdoor temperature.A summary table of the maximum interruptible potential of CRACs under different controllable conditions is given.It is beneficial for data center operators to grasp the interruption potential of CRACs and take advantage of them for grid load regulation. |
PDF Full Text Request