Parametric Reduced-Order Modeling of Aeroelastic Systems

Abstract

In this paper, two approaches for modeling parameter-dependent unsteady aerodynamic loads for control design purposes are presented. The approaches are based on parametric Loewner frameworks, namely, transfer matrix interpolation and global basis. Combined with postprocessing techniques, these frameworks generate highly accurate, reduced-order state-space models of aerodynamic loads. The approaches are computationally efficient since they can model the entire flight envelope while requiring only a single input set of aerodynamic transfer matrix data. Additionally, constant numerical settings can be used across wide ranges of the parameter values without loss of accuracy. The proposed approaches are applied to state-space modeling of a two-dimensional aeroelastic system. The evaluated accuracy and dynamic behavior of the model show excellent agreement with the reference frequency domain solutions for both approaches. https://doi.org/10.1016/j.ifacol.2022.09.087