Fraunhofer Institute for Industrial Mathematics ITWMFeelMath is a fast and easy to use analysis tool for elastic micro-structures given by volume images or analytical descriptions.
FeelMath - Mechanical and Thermal Properties of Microstructures
Extracting Information for the Microstructure
The micro-structure of a material affects its properties like thermal conductivity, elasticity, or acoustic absorption. Image analysis provides micro-structure information needed as input for the dimensioning at the macroscale, i.e. at the level of the whole part.
SEM images, tomographies and the image analysis yield the important information for the micro structure. In addition, the representative volume element is defined small enough not to claim too much capacity but it is also large enough to stay efficiently.
Finding the Optimal Microstructure
Finally, the modeling opens the door to the so-called virtual material design and optimization of micro-structures. Simple changes of model parameters result in slightly altered geometries, in which the target property can be simulated again.
This cycle can be repeated until the optimal micro-structure is found. Thus, costly mechanical tests and production of samples can be reduced while the relation between micro-structure and reaulting properties is better understood.
FeelMathVOX
is used to calculate the effective stiffness of anisotropic, elastic composites and porous materials (e.g., rocks). The big advantage is that the calculation does not require mesh generation, but instead works directly on 3D pixels (voxels).
FeelMathAF
is used to estimate the effective stiffness (directional elastic modulus) of anisotropic, elastic composites with analytical formulas (approximative formulas AF).
FeelMathLD
is used to simulate the physically and geometrically non-linear behavior (large deformations LD) of anisotropic composites and porous materials.
Von-Mises stresses under load in the fiber direction.
Von-Mises stretch under load in z-direction.
Berea sandstone.
Project Example: Integrative Simulation for Structural Components
Fiber reinforced plastics have a high stiffness to weight ratio and can be cost efficiently produced on a mass production scale by injection of compression molding. Therefore, this type of material plays an important role for producing lightweight components. In joint projects with Bosch, we developed an integrative simulation for the dimensioning of short fiber reinforced components, which takes into account the production process as well as the resulting locally varying material properties. During the production process the plastic is injected or molded at medium to high pressure into the component shape. The resulting flow processes are influencing the fiber orientation and thus the mechanical properties significantly.
Fiber reinforced plastics are used in the buckle of a lanyard as well.
Here the boundary conditions of the Abaqus simulation of the buckle are shown.
The various colours show the Von Mises stress obtained by the Abaqus simulation of the buckle.
Integrative Simulation
In the so called off-line phase a material database for different fiber orientations is filled. This stage is relying on a combination of microstructure simulations performed with FeelMath and model order reduction methods. Afterwards, the results of the injection simulation with FLUID, Moldflow or Moldex 3D are transfered onto the Finite Element Mesh (FE-Mesh) of the component simulation.
During the component simulation with Abaqus we are interpolating depending on the actual fiber orientation between the effective material laws obtained during the off-line phase and take account oft he local nonuniform mechanical properties. This allows us to take advantage of the lightweight capabilities of fiber reinforced plastics and to avoid overly large safety factors.
Fiber Reinforced Plastics in the Project MuSiKo
In the research project MuSiko (the abbreviation MuSiKo stands for "Adaptive Approximation Techniques for the Multiscale Simulation of Nonlinear Composite Behavior") we work on the simulation of fiber reinforced plastics as well. We develop efficient multiscale simulation techniques that only need the characteristics and the local orientation of the fibers as input parameters for predicting the macroscopic deformation and failure behavior.