Numerical models for aerospace systems are often of very large dimension. It is not uncommon to solve for millions of degrees of freedom, especially in the field of computational fluid dynamics. The cost in wall-clock time and resources can easily become impractical if the analysis has to be repeated many times, e.g. over the expected flight envelope or atmospheric gust scenarios.
Reduced models are used as acceleration methods to replace/complement the expensive large order models. In this project, reduced models based on system identification methods and manipulation of the original residual will be considered, and the underlying assumptions thoroughly investigated.
The objective is the development and assessment of reduction techniques to drastically reduce the computational cost of simulations for engineering problems. Two approaches are investigated. The first is based on system identification methods, and this work is done in collaboration with the U.S. Air Force Academy of Colorado Springs particularly to model unsteady and non-linear aerodynamic loads. The second approach is based on the manipulation of the original set of equations to speed-up the calculations, and is applicable to any coupled system which is in the form of ordinary differential equations.