Some basic ideas are presented for the construction of robust, computationally efficient reduced order models amenable to be used in industrial environments, combined with somewhat rough computational fluid dynamics solvers. These ideas result from a critical review of the basic principles of proper orthogonal decomposition-based reduced order modeling of both steady and unsteady fluid flows. In particular, the extent to which some artifacts of the computational fluid dynamics solvers can be ignored is addressed, which opens up the possibility of obtaining quite flexible reduced order models. The methods are illustrated with the steady aerodynamic flow around a horizontal tail plane of a commercial aircraft in transonic conditions, and the unsteady lid-driven cavity problem. In both cases, the approximations are fairly good, thus reducing the computational cost by a significant factor.
Reduced order modeling of some fluid flows of industrial interest
Fluid Dynamics Research (Online) ; 44 , 3 ; 031203/1-031203/11
2012
11 Seiten, 43 Quellen
Aufsatz (Zeitschrift)
Englisch
FLUID DYNAMICS - Reduced-Order Modeling of Steady Flows Subject to Aerodynamic Constraints
Online Contents | 2014
|Reduced Order Modeling of Unsteady Flows About Airfoils
British Library Online Contents | 1994
|Reduced-Order Modeling for Non-equilibrium Air Flows
AIAA | 2020
|