Multi-parameter model reduction in multi-scale convective systems

Abstract

A new Proper Orthogonal Decomposition (POD) based reduced order modeling approach for temperature field calculation in multi-scale convective systems is presented. Using POD modes for the entire domain, the energy equation is solved only at the most important scales via Galerkin projection to have an efficient and accurate enough model for design. Comparing with Computational Fluid Dynamics/Heat Transfer simulation shows that the method can predict the temperature field at the rack scale of a sample air-cooled data center with the average error norm of ∼6% for different sets of design parameters, while it can be up to ∼250 times faster.