Numerical investigation of thermal environment in data centers based on fluid flow analysis

With the rapid development of data centers, structural design and energy efficiency optimization have become critical. Airflow organization constitutes the primary determinant of server cabinet thermal environments and cooling system energy consumption. Effective airflow distribution reduces localized hotspots, enhances thermal uniformity, and reduces aggregate energy expenditure. This investigation develops a porous medium model employing simplified cabinet configurations to characterize airflow dynamics in small-to-medium data centers, with particular emphasis on cold-aisle containment efficacy and airflow distribution patterns. Validation metrics demonstrate exceptional model fidelity, with a maximum temperature deviation of 1.42 K (0.45% error) and outlet temperature difference of 1.10 K (0.36% error), confirming the model’s reliability. Numerical simulations and experimental comparisons further verify its feasibility. Using a porous medium model, the study examines the impact of inlet velocity, temperature, server spacing, and height on data center thermal performance. A simplified full-scale data center simulation is conducted to evaluate airflow dynamics under varying load conditions, air supply parameters, and thermal distributions. A parametric analysis establishes the requirements for airflow optimization, as well as the interdependencies between air supply characteristics, cabinet spatial arrangements, and resultant thermal profiles.