Multi-Fidelity Kriging Surrogate Models Applied to the Second UQ Challenge Problem

In this paper, a multi-fidelity kriging surrogate model is employed to tackle the second uncertainty quantification challenge problem where participants are asked to make an estimate of the predictive capability of XFOIL for lift, drag, and moment coefficients at four conditions given ``truth'' (RANS) data in seven different locations (one condition coincides with a truth data point, two are interpolatory, and one is extrapolatory). Accurate surrogate models, once obtained, can be employed for evaluating a large number of designs for uncertainty quantification, optimization, or design space exploration.