SUBJECT: Ph.D. Proposal Presentation
   
BY: Ethan Cruz
   
TIME: Monday, May 6, 2013, 11:00 a.m.
   
PLACE: Love Building, 109
   
TITLE: Viscous-Inviscid Coupled Solution Methodology for Bounded Domains: Application to Data Center Thermal Management
   
COMMITTEE: Dr. Yogendra Joshi, Chair (ME)
Dr. Paul Neitzel (ME)
Dr. Satish Kumar (ME)
Dr. Lakshmi Sankar (AE)
Dr. Paul Krueger (ME)
Dr. Roger Schmidt (IBM)
 

SUMMARY

Air cooling the ever increasing power dissipated from Information Technology (IT) equipment has proved a significant challenge at the data center level. In order to combat this challenge, computational fluid dynamics and heat transfer (CFD/HT) models have been employed as the dominant technique for the design and optimization of both new and existing data centers. Traditional turbulence modeling methods are quite time consuming and surprisingly only accurate in regions of the domain where viscous effects are prevalent. Inviscid modeling has shown great speed advantages over the full Navier-Stokes CFD/HT models (over 20 times faster), but is incapable of capturing the physics in the viscous regions of the domain. A viscous-inviscid coupled solution methodology for bounded domains has been developed in order to increase both the solution speed and accuracy of CFD/HT models. The solution methodology consists of an iterative solution technique that divides the full domain into multiple regions consisting of at least one viscous, inviscid, and interface region. The full steady, Reynolds averaged Navier-Stokes equations with turbulence modeling are used to solve the viscous domain, which consists of the viscous and interface regions. In a similar fashion, the inviscid and interface regions, referred to as the inviscid domain, are solved using the Euler Equations. The two solution domains are solved independently of each other in an iterative manner, and the boundary conditions for each are contained within the interface region of the other. This allows for the boundary conditions to be updated throughout the iterative process. An intermediate step is required in order to ensure the conservation of mass and energy across all boundaries before the updated boundary conditions can be passed from one domain to the other. By combining the increased speed and more accurate results of the inviscid solver in the inviscid regions, along with the viscous solver’s ability to capture the turbulent flow physics in the viscous regions, a faster and overall more accurate solution can be obtained for bounded domains that contain large inviscid regions, such as data centers. Data center CFD/HT modeling will benefit from both the increased accuracy and solution speed with this new inviscid-viscous coupled methodology.