Adaptive Approximation Method for Multi-Scale Simulation of Nonlinear Behavior of Composites

Fiber reinforced plastic components are essential in different applications as e.g. in automotive or medical engineering. In the project MuSiko we develop multiscale simulation techniques for reinforced components.

For microstructured materials, the macroscopic deformation and failure behavior significantly depends on the microstructure, which is influenced by the manufacturing process. For fiber reinforced polymer matrices as for example polybutylenterephthalat (PBT) matrix failure, fiber breaking or delamination may occur. In order to predict these effects often a pure macroscopic analysis is not sufficient. Thus, very complicated phenomenological models are required, which are only valid for certain failure scenarios.


Joint Research Project MuSiKo

In the BMBF joint research project MuSiKo we develop efficient multiscale simulation techniques in consultation with scientists from the universities of Kaiserslautern and Saarbrücken as well as the Karlsruhe Institute of Technology. The abbreviation MuSiKo stands for "Adaptive Approximation Techniques for the Multiscale Simulation of Nonlinear Composite Behavior". The industry partners Robert Bosch GmbH and Siemens PLM Software are supporting the research project.

The applied multiscale approach is based on a coupled solution of the macroscopic and the microscopic problem. Only the characteristics of the matrix the fibers as well as the local orientation of the fibers have to be determined as input parameters for the simulation. The mechanical behavior at the level of component results in the averaged microscopic values.

Process Chain for Glass Fiber Reinforced Plastics

In the project MuSiKo we carried out the complete process chain for glass fiber reinforced PBT  in cooperation with our partners – from the measurement of the plastic properties and the determination of the fiber orientation by means of µCT up to the multiscale simulations. The simulation results are validated via suitable experiments at the component level. With assistance of these simulation techniques, it is possible to optimize the injection molding process (e.g. the temperature or injection point) with regard to the functionality of the components.

Virtuelle Mikrostruktur kurzfaserverstärkter Kunststoff
Virtual microstructure of short fiber reinforced plastics generated with GeoDict software; mass fraction of glass fibers 30%, stress distribution in fibers
Matrix damage, red zones high damage, blue zones low damage
Spannungsverteilung in Filtergehäuse
Stress distribution in ­filter housing of Filtran company under internal pressure, red zones high equivalent stress, blue zones low equivalent stress

Project Partners

  • Rheinland-Pfälzische Technische Universität Kaiserslautern-Landau (RPTU)n (Chair of Technical Mechanics)
  • University of Saarland (Institute of Applied Mathematics and Chair of Technical Mechanics)
  • Karlsruhe Institute of Technology (Institute for Applied and Numerical Mathematics)
  • Robert Bosch GmbH
  • Siemens PLM Software

Summarized Publication

Diebels, Stefan und Sergej Rjasanow (Eds.) (2019):
Multi-scale Simulation of Composite Materials: Results from the Project MuSiKo.
Berlin: Springer-Verlag.