Project Details
Projekt
Data-based scale-bridging simulation of structured magnetorheological elastomers
Applicant
Professor Dr.-Ing. Markus Kästner
Subject Area
Fluid Mechanics
Mechanics
Mechanics
Term
since 2024
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 511114185
The overall objective of this subproject is to investigate the process-structure-property relations of structured magnetorheological elastomers (MREs). In contrast to conventional MREs, they exhibit a pronounced anisotropic arrangement of the contained magnetizable particles in the surrounding soft elastic polymeric carrier matrix. Such structures form, for example, when external magnetic fields are applied during fabrication. In cooperation with the four other subprojects of the research group, the relevant physical processes involved from production to the finished composite material are to be analyzed in a targeted manner using theoretical and experimental methods. In addition, the findings are didactically prepared for high-school students and teachers and are made accessible to a broad public through outreach activities. The focus of this subproject is to provide model- and data-based methods for the efficient cross-scale simulation of structured MREs. Based on the properties of the constituents as well as their microstructural arrangement, the effective macroscopic material behavior is determined by computational homogenization methods. For this purpose, computed tomography (CT) data of real microstructures and data sets from process simulations are statistically analyzed. On this basis, statistically equivalent volume elements (SVEs) with a, compared to the original structure, significantly reduced number of particles are constructed. The advantage of this approach lies in the reduction of the significant computational effort that would be required to simulate the original microstructure. By homogenizing the SVEs, a wide variety of anisotropic structures are investigated with respect to their resulting magneto-mechanical coupling properties. Thus, it is possible to establish the link between process parameters and the effective magneto-mechanical properties of the MREs. Finally, the data set resulting from the simulations serve as a starting point for the formulation and parametrization of a macroscopic constitutive model for structured MREs. For this purpose, a modeling strategy based on artificial neural networks (ANNs) enriched by physical principles is pursued. This approach combines the advantages of classical models with the flexibility of ANNs. The model is then used to study the behavior of complete specimens and actuators for various external loads. Thus, an in-depth understanding of the complex interplay between micro- and macroscopic scales is gained. In the longer term, fully automatic frameworks for autonomous material design will be developed and applied. Furthermore, an extension of models and simulation tools for the representation of dynamic effects is planned.
DFG Programme
Research Units
