Title: Finite Element Updating
Speaker: Richard Ahlfeld, PhD Candidate, Imperial College London
Date: Thursday, September 22, 2016
Time: 3:15pm
Location: NIA, Room 101
Abstract: Finite Element Model Updating is a technique, in which experimentally obtained system response data are used to improve the model’s predictive capability or quantify the structural damage. However, the underlying mathematical problem is very ill-conditioned, non-linear and solutions are not unique. Hence, the updating process is very sensitive to small errors and uncertainties – of which there are unfortunately many. In this work, we try to find out-of-the-box solutions for the two most critical problems that prevent the application of model updating to large FE Models: the incompleteness of either model or measurement data (a large discrepancy between measurement and model) and the difficulty to identify the crucial wrong parameters from a large number (global sensitivity analysis).
Bio: Richard Ahlfeld is a PhD student at the Department of Aeronautical Engineering of Imperial College London. The general focus of his research is on uncertainty quantification and robust design. At Imperial College, he mostly works on Computational Fluid Dynamics (CFD) applications. In the past, however, he has worked on filter and controller design with Airbus Space and on Computer Vision with the German Center of Robotics (DLR).