What is a reduced order model (ROM)?
Definition
A reduced order model (ROM) is a simplified version of a high-fidelity simulation that keeps the dominant behavior of the system while solving far fewer unknowns. Projection-based ROMs extract a small basis from snapshots of full simulations and project the governing equations onto it, so the reduced model can be evaluated cheaply across many parameter values.
Example
In aircraft design, reduced models of coupled aerodynamic and structural behavior are used to evaluate aeroelastic performance across many configurations, where running the full model every time would be too costly. A ROM is one kind of surrogate model: it projects the solver's equations, where data-fit surrogates learn from the solver's inputs and outputs.
Sources
- Benner, Gugercin, Willcox, "A Survey of Projection-Based Model Reduction Methods for Parametric Dynamical Systems," SIAM Review, 2015, PDF
- Beran, Silva, "Reduced-Order Modeling: New Approaches for Computational Physics," AIAA 2001-0853, ntrs.nasa.gov/citations/20010018414