A Bayesian target vector optimization method is fit to a finite element numerical model (JCMsuite) to a Grazing Incidence X-ray fluorescence data set to obtain the geometrical parameters of line gratin within the nanometer range. Convergence studies are performed to determine the numerical parameters that allow for an efficient and accurate reconstruction of the model parameters.
M. Plock, et al., Impact study of numerical discretization accuracy on parameter reconstructions and model parameter distributions. Metrologia, 60, 054001, 2023.