Forward and backward uncertainty quantification with active subspaces: Application to hypersonic flows around a cylinder

Abstract

We perform a Bayesian calibration of the freestream velocity and density starting from measurements of the pressure and heat flux at the stagnation point of a hypersonic highenthalpy flow around a cylinder. The objective is to explore the possibility of using stagnation heat flux measurements, together with pressure measurements, to rebuild freestream conditions since such measurements are available for recent space missions but not exploited for freestream characterization. First, we formulate an algorithm of mesh adaptation, enabling accurate numerical solutions in an automatic way for a given set of inputs. Secondly, active subspaces are used to find a low-dimensional dependence structures in the input-to-output map of the forward numerical solver. Then, surrogate models on the active variables are used to accelerate the forward uncertainty propagation by Monte Carlo sampling and the Markov Chain Monte Carlo sampling of the posterior distribution for Bayesian inversion. A preliminary sensitivity analysis with sparse Polynomial Dimensional Decomposition is performed on the chemical model of the air mixture, to determine the most influential uncertain chemical parameters in the forward problem. Then, the forward and backward methodologies are applied to the simulation of a hypersonic flow around a cylinder, in conditions for which experimental data are available, revealing new insights towards the potential exploitation of heat flux data for freestream rebuilding.

Bayesian Calibration and Sensitivity Analysis for a Karst Aquifer Model Using Active Subspaces

“…This surrogate is constructed according to instructions described in Algorithm 1. There are several examples in the literature that show that a polynomial approximation can be useful in the context of active subspaces, e. g., [15,68]. The accuracy of a regression fit is measured by the r 2 value, or coefficient of determination (see, for example, [25]).…”

On the relation between parameters and discharge data for a lumped karst aquifer model

“…Using these directions, it can also provide global sensitivity metrics for the parameters which is explained below. Recently, active subspaces have been used to reduce the dimension of parameter spaces in high-dimensional Bayesian inverse problems (Constantine et al, 2016;Cortesi et al, 2017;Teixeira Parente et al, 2018.…”

Fixed inducing points online Bayesian calibration for computer models with an application to a scale-resolving CFD simulation

“…For example, Bayarri et al [6][7][8] applied the method to validate models for resistance spot welding, dynamic stress and vehicle collision modelling. In computational fluid dynamic (CFD) simulation, it was applied to tuning the model parameters and inferring the true physical value [9,10], quantifying the inverse uncertainty [11][12][13][14][15][16][17][18] and informing the design of supercomputer simulations [19].…”