Automatic adaptive sampling in parametric Model Order Reduction by Matrix Interpolation

In modeling and simulation of large-scale systems, Model Order Reduction (MOR), and specifically parametric MOR (pMOR), has grown in importance recently. In this paper, a concept based on distance between subspaces has been automated and combined with the efficient parametric reduction approach Matrix Interpolation to give an automatic adaptive sampling strategy in pMOR. An algorithm is developed and its efficacy established with the help of numerical results for a parametric Timoshenko beam model.