These slides were presented at the NAFEMS World Congress 2025, held in Salzburg, Austria from May 19–22, 2025.

Abstract

The production of high-quality foam components for insulation or comfort is a very demanding task, especially when it comes to homogeneous filling of complex assemblies. It is crucial to choose the right amount of injected material, the correct injection position, set accurately thermal conditions, and to place the vents in the proper locations to achieve the desired final product quality. To avoid time-consuming and costly experimental optimization loops, simulation is the smart choice. One crucial aspect is the correct prediction of foam expansion which requires a suitable modelling approach, stable numerical solver and correct model input parameters describing the foaming material. Since foam rheology is a very complex, fulfilling all these parameters is not an easy task. We present a foam simulation technology, addressing both the issues of proper model parameter identification and foam filling in complex parts. We will discuss an automated material parameter identification tool that allows fast material parameterization based on simple foam expansion experiments. The identified parameters are then used directly in the foaming simulations of the complex components and stored in a foam material data base. In addition, we will demonstrate a method to simulate very large and complex parts using an effective porous media approach to accurately resolve small scales. The effective porous media approach substitutes small parts and/or same fine structures by the local permeability. Large parts like battery packs with hundreds of cells can be calculated ten times faster on coarser grids with the same accuracy. Finally, we show how the digital twin can be used to predict and control the thermo-mechanical product quality of foam components. The digital twin begins by simulating the foaming process to determine the local density and pore size distribution of the foam component. Based on this, a foam database for different densities and pore sizes is dynamically created. This step relies on microstructure simulations to determine the effective thermo-mechanical behaviour of the foam for every density. Using the data base results from the microstructure simulation together with the local information from foam process simulation, the component design of the foam parts can be optimized using a standard FE-tool, considering the different local material properties, which depend on the process conditions . Some simulations of various applications, including the isolation of battery packs or the foaming of textiles to produce reinforced lightweight structure as well as refrigerators, seats or insulation panels will be presented and discussed to showcase the capabilities of digital prediction software.