Dr. Bassam Khuwaileh

Research Associate at Los Alamos National Laboratory (CCS-6 group)

Abstract

Nuclear scientists and engineers use computer models to study and simulate parameters-responses relationship.However, thorough parameter-response studies are challenging and  sometimes impossible when the simulation model is computationally expensive and/or the number of model parameters/responses is relatively large. To enable such studies, this talk proposes the use of the so called active subspace and Reduced Order Modeling techniques to tackle this problem.
Reduced Order Modeling refers to the set of mathematical techniques that are used to reduce the computational burden of conventional modeling techniques, such as finite element and finite difference methods, while minimizing the loss of accuracy that typically accompanies such a reduction. Aided with the subspace analysis, ROM techniques can offer nuclear engineers with a powerful tool for forward and inverse analysis. This talk will overview and present few case studies that resembles the use of subspace based ROM techniques in both forward and inverse analysis.

Bio of Dr. Khuwaileh

Dr. Khuwaileh is currently a research associate at Los Alamos National Laboratory (CCS-6 group). Dr. Khuwaileh received his PhD in Nuclear Engineering from
North Carolina State University (2015), after which he joined the CCS-6 group at Los Alamos National Laboratory to work on developing and implementing state of art uncertainty/sensitivity and model calibration methodologies for reactor physics calculations.