Colloquia

Samenvatting

The numerical modelling of the constitutive response of materials is important in a wide range of applications. With the growing market interest in electrical machines it becomes inherent that models predicting material behavior of parts used in these machines are gaining attention, too. As most material properties, influencing whether a material is suited for the use in an electrical machine, are defined on the atomic length-scale, modern finite element methods are needed for these predictions. The most commonly applied method is crystal plasticity finite element. Within this work a novel crystal plasticity framework, being in theory more robust and stable than conventional frameworks, is validated and optimized for the use of predicting material behavior of steel used in induction motors. A proper element type is chosen and a link between crystallographic descriptions of steel and the commercial finite element solver Abaqus is constructed. Then, a three dimensional model of a micro tensile steel specimen is generated and simulated. Results show that the crystal plasticity framework is able to predict the inhomogeneity of plastic deformation in this particular steel. This inhomogeneity was validated by comparisons to experimental work found in scientific literature. Further, experimental validation of the numerical data was realized by measuring the crystal orientations of a steel sample. The steel sample was deformed and the orientations were measured again. The same procedure was done on numerical level, showing similarities and differences in the rotation of crystals during deformation. Finally, a tensile-compression asymmetry of crystal rotations was shown on numerical level. Results indicate that applying a compressive load will cause stronger rotations of the crystals. This relates back to the observation that certain crystal-orientation-dependent properties of steel deteriorate stronger when the material is subjected to compression than tension. Summarized, the novel crystal plasticity finite element framework is capable of predicting the material behavior of steel under different loading conditions. The models used in this work are large scale 3D representations of real life experimental specimens. The complexity of these models is high compared to currently reported approaches in literature on this topic. Implementing the aforementioned improvements, the novel crystal plasticity finite element framework holds great potential for future application for the prediction of material behavior.