Efficient Reduced-Order Modeling of Unsteady Aerodynamics under Light Dynamic Stall Conditions

Li Xiaohua,Zheng Guo,Grecov Dana,Zhongxi Hou

DOI: https://doi.org/10.1177/0954410018773628

2019-01-01

Abstract:In this research, a reduced-order modeling is developed to predict the unsteady aerodynamic forces under light dynamic stall conditions at low-speed regimes. The filtered white Gaussian noise is selected as input signals for computational fluid dynamics solver in order to generate training data, containing the information of reduced frequency and amplitude. Because of the time history influences, the reduced-order modeling combines the Kriging function and recurrence framework together in this approach. An airfoil NACA0012 undergoing pitching motions with different reduced frequency, amplitude, and mean angle of attack is designed to illustrate the methodology. The developed model can predict the lift, drag, and moment coefficients in seconds on a single-core computer processor. To reduce the prediction errors between reduced-order modeling predictions and computational fluid dynamics simulations, the aerodynamic loads in static conditions are applied as initial inputs. The predictions via the proposed approach are in agreement with the results using a high precision computational fluid dynamics solver over the designed ranges of amplitude and reduced frequency, which is suitable for engineering applications, such as fluid-structure interaction, and aircraft design optimizations.