@inproceedings{HollweckLeidingerHartmannetal., author = {Hollweck, Christoph and Leidinger, Lukas and Hartmann, Stefan and Li, Liping and Wagner, Marcus and W{\"u}chner, Roland}, title = {Systematic assessment of isogeometric sheet metal forming simulations based on trimmed, multi-patch NURBS models in LS-DYNA}, series = {14th European LS-DYNA Conference, October 18 and 19, 2023, Baden-Baden, Germany}, booktitle = {14th European LS-DYNA Conference, October 18 and 19, 2023, Baden-Baden, Germany}, publisher = {DYNAmore}, doi = {10.35096/othr/pub-6822}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:898-opus4-68228}, abstract = {Isogeometric sheet metal forming simulation is a numerical simulation technique that is used to predict the behavior of sheet metal parts during the forming process [1] and tries to tighten the link with the Computer Aided Design (CAD) description. This technique uses the isogeometric analysis (IGA) approach, which combines the well-established framework of traditional finite element analysis (FEA) and the power of non-uniform rational B-splines (NURBS). In contrast to the approach in a "classical" FEA framework, IGA directly uses the ansatzspace of the CAD geometry for analysis, which opens the possibility to work directly on the exact geometry. Furthermore, the smoothness of the NURBS basis functions results in a more accurate simulation [2]. A powerful method to reduce the computational effort is adaptive mesh refinement, that has been developed and optimized for sheet metal forming applications over several years for standard Finite Elements. However, it remains an open question how an efficient local adaptive mesh refinement strategy can be implemented for complex industrial sheet metal forming simulations based on trimmed NURBS models, which are typically the description in Boundary Representation (B-Rep) CAD-models [3]. First investigations for explicit dynamics have been made in [4]. In this contribution, a detailed comparison between FEA and IGA sheet metal forming applications is conducted. The state of the art for FEA and IGA will be contrasted and the need for an efficient adaptive mesh refinement strategy will be discussed. The goal of our research is to develop an efficient adaptive mesh refinement strategy for isogeometric sheet metal forming simulations in LS-DYNA. This will contribute to closing the efficiency gap between IGA and FEA in explicit dynamics, accelerate the product development process and enable the application of IGA in industrial sheet metal forming simulations.}, language = {en} } @article{KapsLehrerLepeniesetal., author = {Kaps, Arne and Lehrer, Tobias and Lepenies, Ingolf and Wagner, Marcus and Duddeck, Fabian}, title = {Multi-fidelity optimization of metal sheets concerning manufacturability in deep-drawing processes}, series = {Structural and Multidisciplinary Optimization}, volume = {66}, journal = {Structural and Multidisciplinary Optimization}, publisher = {Springer Nature}, doi = {10.1007/s00158-023-03631-8}, abstract = {Multi-fidelity optimization, which complements an expensive high-fidelity function with cheaper low-fidelity functions, has been successfully applied in many fields of structural optimization. In the present work, an exemplary cross-die deep-drawing optimization problem is investigated to compare different objective functions and to assess the performance of a multi-fidelity efficient global optimization technique. To that end, hierarchical kriging is combined with an infill criterion called variable-fidelity expected improvement. Findings depend significantly on the choice of objective function, highlighting the importance of careful consideration when defining an objective function. We show that one function based on the share of bad elements in a forming limit diagram is not well suited to optimize the example problem. In contrast, two other definitions of objective functions, the average sheet thickness reduction and an averaged limit violation in the forming limit diagram, confirm the potential of a multi-fidelity approach. They significantly reduce computational cost at comparable result quality or even improve result quality compared to a single-fidelity optimization.}, language = {en} } @article{LehrerKapsLepeniesetal., author = {Lehrer, Tobias and Kaps, Arne and Lepenies, Ingolf and Duddeck, Fabian and Wagner, Marcus}, title = {2S-ML: A simulation-based classification and regression approach for drawability assessment in deep drawing}, series = {International Journal of Material Forming}, volume = {16}, journal = {International Journal of Material Forming}, publisher = {Springer}, doi = {10.1007/s12289-023-01770-3}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:898-opus4-64664}, pages = {1 -- 17}, abstract = {New structural sheet metal parts are developed in an iterative, time-consuming manner. To improve the reproducibility and speed up the iterative drawability assessment, we propose a novel low-dimensional multi-fidelity inspired machine learning architecture. The approach utilizes the results of low-fidelity and high-fidelity finite element deep drawing simulation schemes. It hereby relies not only on parameters, but also on additional features to improve the generalization ability and applicability of the drawability assessment compared to classical approaches. Using the machine learning approach on a generated data set for a wide range of different cross-die drawing configurations, a classifier is trained to distinguish between drawable and non-drawable setups. Furthermore, two regression models, one for drawable and one for non-drawable designs are developed that rank designs by drawability. At instantaneous evaluation time, classification scores of high accuracy as well as regression scores of high quality for both regressors are achieved. The presented models can substitute low-fidelity finite element models due to their low evaluation times while at the same time, their predictive quality is close to high-fidelity models. This approach may enable fast and efficient assessments of designs in early development phases at the accuracy of a later design phase in the future.}, language = {en} } @misc{LehrerStockerDuddecketal., author = {Lehrer, Tobias and Stocker, Philipp and Duddeck, Fabian and Wagner, Marcus}, title = {Comparison of Low- vs. High-Dimensional Machine Learning Approaches for Sheet Metal Drawability Assessment}, series = {Third International Conference on Computational Science and AI in Industry (CSAI 2023), Trondheim, Norway, 28-30 August 2023}, journal = {Third International Conference on Computational Science and AI in Industry (CSAI 2023), Trondheim, Norway, 28-30 August 2023}, publisher = {International Centre for Numerical Methods in Engineering}, doi = {10.35096/othr/pub-6477}, url = {http://nbn-resolving.de/urn:nbn:de:bvb:898-opus4-64772}, abstract = {Developing new deep-drawn sheet metal parts is a challenging task due to conflicting demands for low costs, durability, and crash properties. Ensuring manufacturability throughout geometrical changes adds to the complexity, leading engineers to rely on experience-driven iterative design changes that compromise requirements and lack reproducibility. Finite Element (FE) simulation models are employed to ensure manufacturability, albeit at the expense of high computational costs and delays in part development. To improve efficiency, a Machine learning (ML)-centered approach was proposed to ensure manufacturability. However, the limited availability of data raises uncertainty about whether a low- or high-dimensional ML approach is most suitable for drawability assessment. This work compares the accuracy of a low-dimensional, feature-based Linear Support Vector surrogate and an adapted high-dimensional PointNet model under different dataset sizes. The dataset is composed of parametrically generated, U-shaped structural sheet metal parts. We use a one-step simulation scheme and evaluate results with a Forming Limit Diagram (FLD) to label drawability. Results show the point of transition to be at about 500 training samples, from which onwards Deep learning is advantageous. Moreover, the generalizability of these models is tested on a second dataset with topologically similar components. This is to assess the potential for a geometrically more comprehensive evaluation. We discuss several influences on model performances and outline future potentials.}, language = {en} } @article{ThumannBuchnerMarburgetal., author = {Thumann, Philipp and Buchner, Stefan and Marburg, Steffen and Wagner, Marcus}, title = {A comparative study of Glinka and Neuber approaches for fatigue strength assessment on 42CrMoS4-QT specimens}, series = {Strain}, volume = {2023}, journal = {Strain}, number = {e12470}, publisher = {Wiley}, issn = {1475-1305}, doi = {10.1111/str.12470}, pages = {21}, abstract = {In fatigue strength assessment, the methods based on ideal elastic stresses according to Basquin and the less established method based on elastic-plastic stress quantities according to Manson, Coffin and Morrow are applied. The former calculates loads using linear-elastic stresses, the latter requires elasticplastic evaluation parameters, such as stresses and strains. These can be determined by finite element analysis (FEA) with a linear-elastic constitutive law, and subsequent conversion to elastic-plastic loads, using the macro support formula by Neuber. In this contribution, an alternative approach to approximate elastic-plastic parameters proposed by Glinka is compared to the the strain-life method using Neuber's formula, as well as the stress-life method of Basquin. Several component tests on 42CrMoS4-QT specimens are investigated. To determine the input data for the fatigue strength evaluations, the entire test setup is computed by FEA. The nodal displacements from these validated full-model simulations are used as boundary conditions for a submodel simulation of a notch, whose results serve as input for the fatigue strength assessments. It is shown that all approaches provide a reliable assessment of components. Our key result is that the strain-life method using the concept by Glinka for notch stress computation, yields improved results in fatigue strength assessments.}, language = {en} }