The friction welding is a well-established industrial joining process because it is economical, robust, precise, and reproducible.
The reasons for simulating the friction welding process are manifold in scientific and industrial contexts. It ranges from a general increase in understanding of process physics and the prediction of residual stresses to the optimization of process parameters and the development of new friction welding process variants.
Against this background, a number of different simulation approaches have been published, which differ mainly in the modeling of material behavior, friction, contact, and the solution methodology for the thermomechanical problem. A drawback of these FE-simulation approaches is that local, large deformations cannot be taken into account, or can only be considered within the context of a re-meshing of the geometries during time integration. However, in addition to the obvious increase in time and computing effort required for the discretization process itself, the re-meshing also causes significant additional effort in the solution process due to the repeated creation and inversion of the tangent matrix.
To overcome these disadvantages, an alternative approach to the usual isoparametric finite elements can be used, which uses two different types of shape functions to approximate the virtual and non-virtual field variables. As a result, an asymmetric element stiffness matrix is obtained, which is why the method is also referred to as the unsymmetric finite element method (UFEM).
It can be shown that this alternative formulation is mesh distortion resistant, even with large deformations and highly distorted elements, accurate results are generated. In this way, it is possible to avoid the problem of repeated remeshing when simulating the friction welding process, which promises significant advantages in terms of both computing time and post-processing.
Despite the development of the UFEM since the 2000s, its application to nonlinear problems remained insufficient.
As part of the project, the application of UFEM for friction welding simulations will therefore be developed and analyzed in detail, taking into account the process-specific conditions with regard to the advantages and potential disadvantages.

View project in the research portal

The declared aim of the project is the continuous implementation of the models created in the project: Friction-welded hybrid joints made of aluminum and steel: experimental investigation and phenomenological modeling. A simulation platform is being developed specifically for this purpose, in which the calculations for the process, material and structural simulation (virtual tensile test) converge incrementally. The modeling method is then critically evaluated by validating the simulated load-bearing capacity with corresponding experimental data. After successful validation, the load-bearing capacity of the hybrid joint is to be improved through targeted process optimization.
This text was translated with DeepL

View project in the research portal

Friction welding is an established joining process that is used in many areas of mechanical engineering to produce hybrid structures made of aluminum and steel. The decisive factor for the serviceability of hybrid joints is above all the material-appropriate design of the joint. Due to the dependence of the welded joint on the shape, type and continuity of the intermetallic diffusion layer, the microstructure and the material bonding, the development of friction-welded hybrid structures with optimum properties is often time-consuming and cost-intensive. For kmU in particular, it is therefore almost impossible to develop such hybrid structures economically. The declared aim of the project is to set up and test a simulation for the design of friction-welded hybrid joints made of aluminum and steel. For this purpose, corresponding friction welding tests are carried out, whereby the process parameters are systematically varied. These tests provide the data basis for the experimental analysis of the macroscopic, mesoscopic and microscopic influences on the load-bearing capacity of the structure. At the same time, the tests serve as a validation basis for the simulation of the welding process itself. With the help of the process simulation, the effects of the process parameters on the process variables and thus on the material and structural properties can be derived. Based on this, corresponding phenomenological models are developed in order to map the relevant influences. These results are then used as initial conditions for the simulation of the load-bearing capacity (virtual tensile test) of the hybrid joint.
For kmU in particular, the simulation creates the economic possibility of predictively evaluating the joint depending on the selected process. Complex friction welding tasks can be analyzed in advance of the test and optimized accordingly.
This text was translated with DeepL

View project in the research portal

The aim of the project is to develop an externally toothed gear wheel in an Al-Fe mixed construction. The inner part of the gearwheel is to be made of aluminum and welded into the steel outer ring. A friction welded joint is to be realized on the circumferential surface to ensure the safe transmission of the highest possible forces and torques. Predictive methods of friction welding simulation are used for the design of the joint geometry, which enable the process parameters to be determined, on the basis of which a multi-criteria optimization of the joining process can be carried out.
This text was translated with DeepL

View project in the research portal