Towards a predictive crash simulation of composite structures via high fidelity finite element models using high-performance computing

Within the scope of a joint basic research project, the Korea Institute of Material  Science (KIMS) and the Institute of Lightweight Engineering and Polymer Technology (ILK) investigate the crash behaviour of a novel thermoplastic fibre composite material. In this context, tubular test specimens are investigated experimentally under axial compressive loading and numerically on high-performance computing facilities at the Centre for Information Services and High Performance Computing (Zentrum für Informationsdienste und Hochleistungsrechnen (ZIH)) at TUD Dresden University of Technology.

In contrast to metals, which absorb energy through plastic deformation and progressive folding, thermoplastic composite materials show very different energy absorption mechanisms, based mostly on material failure. These failure phenomena significantly affects the overall structural failure pattern and crashworthines of the composite structure. Conventional finite element modelling strategies lack of sufficient representation of these material specific phenomena leading to large uncertainties and prevent wider use of composites in crash applications.

In respect to a sufficient accurate simulation of the crash behaviour, a multi-layered high-fidelity solid modelling approach is applied taking into account every single ply and interfaces of the composite layup. Current increasing computing capacities enable the solution of such high-fidelity models and the adequate representation of both all relevant local deformation and damage phenomena under crash loading and the resulting force-displacement response of the structure.

Our results are recently puplished online open access in the journal Composite Structures. This article is an outcome of the computational methods and simulation specialist group around Holger Böhm, Jonas Richter and Dr. Andreas Hornig supervised by Prof. Maik Gude from the ILK and Dr. Jinbong Kim from KIMS.

https://doi.org/10.1016/j.compstruct.2023.117115

Contact persons:
Dipl.-Ing. Holger Böhm, Expert Group: Computational Methods and Simulation, +49 351 463 38019,
Dipl.-Ing. Jonas Richter, Expert Group: Computational Methods and Simulation, +49 351 463 38332,