Research on Model Validation and Uncertainty Quantification of Airfoil Flow

Abstract

Abstract Numerical simulations and wind tunnel tests are both basic methods for aerodynamic research and application. They are the modeling and simulation of real world system. There are various uncertainties in the implementation process. In Computational Fluid Dynamics (CFD), the main sources of uncertainty are the model form, model parameters and numerical solution methods. In wind tunnel tests, there are many uncertainties sources, such as geometric model, test environment, test devices and instrumentation, test methods, data recording and processing, etc. This paper takes NACA0012 airfoil as an example. Bayesian model averaging method (BMA) and grid convergence study are used to analyse model form and numerical solution uncertainty in CFD. For the geometric configuration uncertainty in wind tunnel test, regression analysis and bootstrap method are used to carry out data fusion. Finally, model validation research is conducted.