Unsteady Aerodynamics Modeling for Aircraft Maneuvers: a New Approach Using Time-Dependent Surrogate Modeling

Abstract

A new approach for computing the unsteady and nonlinear aerodynamic loads acting on a maneuvering aircraft is presented. This approach is based on Duhamel’s superposition integral using indicial (step) response functions. The novelty of this approach relies on the development of a time-dependent surrogate model that fits the relationship between flight conditions (Mach number and angle of attack) and indicial functions calculated from a limited number of simulations (samples). The aircraft studied in the current paper exhibit highly nonlinear roll moment and therefore a very large number of step functions need to be calculated to accurately predict the aerodynamic behavior at each instant of time spent in aircraft maneuvers. The reduced order model, along with the surrogate model, provide a mean for rapid calculation of step functions and predicting aerodynamic forces and moments during maneuvering flight. The maneuvers are generated using a time-optimal prediction code with the feasible solutions based on the vehicle control and state constraints. Results presented show that the developed surrogate model aids in reducing the overall computational cost to develop cost-effective reduced-order models. It is also demonstrated that the reduced order model used can accurately predict timemarching solutions of maneuvering aircraft, but with an advantage that reduced order model predictions only require on the order of a few seconds of computational time.