Research projects
Research topics currently being worked on at the Chair of Mechanics of Multifunctional Structures:
Joint project of the University of Paderborn, the TU Dresden and the Friedrich-Alexander University Erlangen-Nuremberg
Sub-project B03: Corrosion modeling for the evaluation of mechanically joined components
In modern lightweight structures in mixed construction, corrosion, especially contact corrosion, e.g. between steel and aluminum, plays an important role in the evaluation of the load-bearing capacity due to the large number of material combinations, since corrosion can significantly reduce the fatigue strength. In addition to the joinability prediction, the prediction of the service life of the joint, particularly in corrosive media, is another requirement to be met for the increased use of mechanical joining processes. For mechanical joining to be applicable as a versatile process, it must be possible to reliably map the entire process chain from manufacture to operation in virtual product development, because only then can the integrity of the assembly be guaranteed for the entire service life without the need for time-consuming tests. Damage mechanics has established itself as a powerful method which, on the basis of coupled simulations, can supplement or eliminate the need for complex experimental component testing in conjunction with classic fatigue strength concepts. Corrosion describes
describes the change in material as a result of electrochemical reactions, which can have a wide variety of causes. It can be described using physically motivated models based on transport equations for the particle types involved and corresponding approaches for the reaction kinetics. The material properties change as a result of the chemical reactions. This material change must be taken into account in the damage model. In order to make this behavior accessible to simulation, a coupled electro-chemo-mechanical model is to be created in this sub-project, with which the service life of components joined by mechanical joining processes can be predicted under cyclic operating loads and the influence of corrosion up to cracking.
Improving the transfer(mapping) of simulation data from forming simulationto crash simulationfor more accurate failure predictions - " MapUmCra"
In connection with the ever-increasing demand for lightweight construction to reduce the CO2 footprint, the most accurate possible prediction of component behavior in use is essential in the field of vehicle development, for example in crash structure design. In this area, SMEs are usually active as simulation service providers, suppliers of subcomponents and software developers and must adapt to the increasing requirements of various OEMs in order to remain competitive. In order to enable fast, efficient product development, the continuous virtual mapping of the process chain from the forming process to functional use (e.g. crash loading) is used. However, a wide range of incompatibilities arise here, which cause fundamental problems, as different deformation and damage models, solvers, element formulations and mesh refinements are usually used, and these differ between SMEs and OEMs. Load path changes and different load velocities in the forming and crash process also pose further challenges. In this project, systematic experimental and numerical investigations are therefore to be carried out for a steel and aluminum sheet material as an example in order to determine the relevant influencing variables from the forming process that significantly affect crash safety. Based on this, an effective procedure for the transfer (mapping) of the variables identified as relevant from the forming process into the crash simulation for frequently used deformation and damage models as well as different mesh sizes is to be developed and validated using sample components. The interface between the forming and crash simulation with different material models and simulation software solutions will be designed to be interoperable, which will bring great advantages, especially for SMEs that serve the entire process chain.
in the DFG priority program 1681 "Field-controlled particle-matrix interactions: Generation, cross-scale modeling and application of magnetic hybrid materials"
Due to their stimuli-responsive properties, multifunctional materials are predestined for actuator and sensor applications, e.g. in microfluidics and medical technology. Important representatives here are polymers that are sensitive to externally applied electric and magnetic fields and enable controlled deformation.The present application aims to investigate in detail the behavior of ferrogels, i.e. polymer gels with magnetic particles, under the influence of an external magnetic field. For this purpose, the (chemo-)magneto-mechanical behavior of the ferrogels is modeled and simulated using finite elements. The ferrogel consisting of a solid polymer network, a liquid phase and fixed or mobile magnetic particles is described on the basis of a theory of porous media for the different phases. The overall system is described by coupling the mechanical and magnetic field equations. The parameters for the material model are obtained from microscopic experimental and theoretical/numerical investigations by project partners. Within the project, the influence of the external magnetic field on the magnetic particles and the interaction of the particles with the polymer network and the resulting mechanical deformation of the gel will be investigated. For isotropic and anisotropic gels - with fixed or mobile magnetic particles - the behavior in the magnetic field will be calculated and compared with results from experimental working groups. First, the magneto-mechanical hysteresis behavior of ferrogels will be analyzed and taken into account in the developed model. Furthermore, mobile electrically charged particles will also be taken into account. Thus, the deformation behavior of polymer gels under electric and magnetic fields can be determined by numerical simulations and the quality of the developed chemo-electro-magneto-mechanical formulation can be continuously improved in coordination with experimental investigations within the SPP. A thermodynamically consistent model will then be available as a tool for modeling and investigating ferrogel actuator configurations relevant to medical technology and microfluidics.
The project aims to advance the development and optimization of Solid Oxide Fuel Cells (SOFCs) through a hybrid modeling framework that combines physics-based simulations with machine learning (ML) methods. The starting point is the creation of a structured multiphysics, coupled model at the fuel cell level that captures transport processes, electrochemical reactions, thermal effects, and mechanical loads. Building on this, homogenization approaches will be employed to determine effective temperature-dependent material properties of electrodes derived from tomography data and artificially generated microstructures, and to integrate them into macroscopic simulations.
Another key focus is the development of a framework that links microstructural features (e.g., porosity, tortuosity, triple-phase boundaries) to macroscopic transport properties. The generated datasets will be used to train neural networks, enabling fast and real-time predictions and optimizations. In addition, the inclusion of mechanical degradation mechanisms such as creep behavior will allow for long-term reliability forecasts.
To ensure data interoperability and knowledge integration, an ontology-based platform (EMMO) will be applied to support the development of a “material twin” for SOFCs. Overall, the project aims to develop more accurate, scalable, and efficient simulation and prediction methods for SOFCs, thereby laying the foundation for high-performance, sustainable energy solutions.
Joint project with Prof. Dr.-Ing. Berthold Schlecht from the Chair of Machine Elements
and Dr. E.-F. Markus Vorrath from the Chair of Microsystems
Funded from 2021-2026
In the course of Industry 4.0, there has been a significant increase in the demand for data. This data is essential for the implementation of predictive maintenance. The use of a large number of sensors is required to collect this process data. In this case, it is advisable to measure the respective data directly at the point of origin, such as the transmitted torque of a clutch.
For this purpose, a spider coupling is equipped with dielectric elastomer sensors as part of the project. With the help of these sensors, it is possible to measure the deformation of the coupling's spider. This allows conclusions to be drawn about the transmitted torque.
The main focus of the Chair of Mechanics of Multifunctional Structures is on (i) modeling and numerical simulation of the complex thermo-mechanical material behavior of the coupling, (ii) investigating whether damage can occur due to sensor integration and (iii) creating a suitable regression model to determine the torque based on the measured capacities and temperatures.
Further information regarding the priority program can be found on the following website.