Comparison and Application of Different Viscous Models for Indoor Air Flow Simulation
Abstract
Indoor air flow is a term referring to the thermal and fluid flow in enclosed environments. It is closely related to the human and the stable operation of a data center. Computational fluid dynamics (CFD) has played an important role in designing and evaluating various indoor air distributions. To determine which viscous model works best for indoor environment, six commonly used viscous models are evaluated. They are laminar model, standard k-ϵ model, renormalization group (RNG) k-ϵ model, realizable k-ϵ model, standard k-ω model and shear-stress transport (SST) k-ω model. Simulation results are compared with experimental data for three typical indoor air flow patterns: natural convection in a tall cavity, forced convection in a model room with partitions and mixed convection in a square cavity. None of the viscous models have the optimized performance for all types of flow. The overall performance of the RNG k-ϵ model is better than others and is therefore recommended for simulations of indoor air flow. The RNG k-ϵ model is then used to simulate the air distributions for the data center at Purdue University Calumet. Based on the analysis of the thermal and fluid flow field, the optimal configuration for the data center is proposed.
Degree
M.S.E.
Advisors
Wang, Purdue University.
Subject Area
Mechanical engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.