Model Reduction of Inhomogeneous Initial Conditions

Caleb Magruder

Our goal is to develop model reduction processes for linear dynamical systems with non-zero initial conditions. Standard model reduction schemes optimize system input/output characteristics but frequently destroy initial condition state space information. Additionally, existing model reduction schemes for initial conditions inappropriately assign weight to initial condition information.We propose a projection-based model reduction scheme that combines subspaces from differing model reduction approaches and compare it to existing schemes.