Although full-field simulation using computational fluid dynamics and heat transfer (CFD/HT) tools can be applied to predict the flow and temperature fields inside data centers, their running time remains the biggest challenge to most modelers. From a simulation standpoint, there are still rooms for improving the speed of a full-field simulation process of a CFD/HT model whose bounded domain mostly consists of inviscid regions such as data center. Since the inviscid domain is mainly solved using Euler equations, the computing time for it is much faster than solving full Navier-Stokes equations with turbulent models for viscous domains. However, it is less desired to fully replace the viscous regions due its incapability of capturing the physics in these regions such as turbulence. Therefore, if inviscid domain is solved simultaneously with viscous domain, both the speed and accuracy will be much improved. In this paper, the inviscid-viscous coupling strategy is introduced in the solution domain to drastically reduce the running time while preserving the accuracy of a data center model. New criteria for dividing the inviscid, viscous regions, as well as interface region are presented. Step by step instruction on how to construct such regions for a data center model is also provided. The results show this approach’s superb simulation speed, while the accuracy is mostly retained from a full CFD/HT simulation.
See more of: Airflow Requirements and Modeling Approaches