Kriging-based gust response analysis methodology at high angles of attack

The present study demonstrates a kriging-based approach for gust response analysis at high angles of attack. Such flight conditions lead to flow separation and require high-fidelity computational fluid dynamic (CFD) simulations for accurate prediction of the gust loads. However, the computational expense associated with CFD simulations limits their applicability for design applications. A kriging-based interpolation methodology can offer a computationally inexpensive alternative without adversely affecting the accuracy of the predictions. Once the parameter space used for computing the gust responses is sampled appropriately, kriging can interpolate the fully nonlinear gust responses obtained from EZNSS, a turbulent unsteady Reynolds-Averaged Navier-Stokes code, to predict the responses at the design points. The present article provides a brief description of the kriging approach and the kriging-based methodology. The validation of the kriging-based methodology for predicting aeroelastic responses to one-minus-cosine gust excitations at is presented here. At incipient stall conditions, the kriging-based approach could predict peak generalized modal amplitude within 12% of the CFD-based responses and peak phase within 2% of the CFD –based predictions. Overall, the kriging-based predictions showed a similar behavior of the peak gust response from CFD-based predictions for a wide range of gust gradient lengths.