Visiting Research Scientist
1) Efficient uncertainty quantification schemes for handing large number of uncertain variables Most approaches to deal with the situation with large number of stochastic variables are based on sampling strategies (e.g sparse sampling techniques). An alternative method is to look at the reduced basis decompositions for the stochastic fields. At VUB, we developed a non-intrusive reduced basis method for handing large number of uncertainties efficiently. The main idea is to extract the optimal orthogonal basis via inexpensive calculations on a coarse mesh and then use them for the fine scale analysis.
2) Robust optimization by combination of adjoint methods with polynomial chaos expansion Robust design is a design methodology for improving the quality of a product by minimizing the impact of uncertainties on the product. The goal at VUB is to build foundations for combination of uncertainty quantification with the adjoint to form an efficient computational framework for the robust optimization.