Computational Homogenization of DP Steels using Statistically Similar RVEs
Event details
| Date | 05.03.2013 |
| Hour | 13:15 › 14:15 |
| Speaker |
Dr.-Ing. Daniel Balzani Bio : Dr. Daniel Balzani is senior research assistant at the Institute of Mechanics, Faculty of Engineering, Department of Civil Engineering, University of Duisburg-Essen. He received his diploma of engineering science there in 2003 and his doctoral degree from the Technical University of Darmstadt in 2006. Since then he is working in Essen again with an interruption in 2009 when he worked for one year at the Leibniz University Hannover as a Substitute Professor for Mechanics at the Institute of Mechanics and Computational Mechanics (IBNM) and in 2010 when he joined the research group of Michael Ortiz at the California Institute of Technology in Pasadena as visiting faculty for seven months. His habilitation thesis was successfully defended in December 2012. Daniel Balzani is currently PI of five projects and author or co-author of 17 peer-reviewed journal articles and over 40 conference proceedings. His research interests are in the field of continuum mechanical modeling of nonlinear micro-heterogeneous materials and numerical methods for their simulation. |
| Location | |
| Category | Conferences - Seminars |
Abstract : In many applications in structural engineering advanced high strength steels are more and more important due to their enhanced properties of increased stability and formability. These properties originate from the interaction of heterogeneously distributed constituents on the microscale leading to a complex elasto-plastic material behavior at the macroscale including kinematic hardening. A numerical method to directly incorporate the microscopic mechanical fields into the simulation are direct micro-macro-transition approaches, such as the FE² method, where in each macroscopic evaluation point an additional boundary value problem on the microscale is solved and the macroscopic quantities are calculated based on suitable volume averages of microscopic quantities. The microscopic boundary value problem is constructed based on a representative volume element (RVE) which leads to high computational costs for RVEs that are constructed as substructures of real microstructures. The high complexity of the morphology of such RVEs results in a high number of degrees of freedom in the RVE. This drawback can be circumvented by the usage of statistically similar RVEs (SSRVEs), which have the ability to increase the efficiency of the method. SSRVEs are characterized by a lower complexity than RVEs while giving a rather accurate representation of the mechanical response of the real microstructure. For the construction of SSRVEs a least-square functional is minimized considering the differences of suitable statistical measures calculated for the real microstructure and the SSRVE. The main advantage is that thereby complex micro-macro calculations can be performed where mechanical phenomena such as e.g. eigenstrains can also be incorporated in a direct sense even at the microscopic level.
Practical information
- General public
- Free
Organizer
- IGM
Contact
- Géraldine Palaj