Print Email Facebook Twitter Physically recurrent neural networks for path-dependent heterogeneous materials Title Physically recurrent neural networks for path-dependent heterogeneous materials: Embedding constitutive models in a data-driven surrogate Author Alves Maia, M. (TU Delft Applied Mechanics) Rocha, I.B.C.M. (TU Delft Applied Mechanics) Kerfriden, P. (PSL Research University) van der Meer, F.P. (TU Delft Applied Mechanics) Date 2023 Abstract Driven by the need to accelerate numerical simulations, the use of machine learning techniques is rapidly growing in the field of computational solid mechanics. Their application is especially advantageous in concurrent multiscale finite element analysis (FE2) due to the exceedingly high computational costs often associated with it and the high number of similar micromechanical analyses involved. To tackle the issue, using surrogate models to approximate the microscopic behavior and accelerate the simulations is a promising and increasingly popular strategy. However, several challenges related to their data-driven nature compromise the reliability of surrogate models in material modeling. The alternative explored in this work is to reintroduce some of the physics-based knowledge of classical constitutive modeling into a neural network by employing the actual material models used in the full-order micromodel to introduce non-linearity. Thus, path-dependency arises naturally since every material model in the layer keeps track of its own internal variables. For the numerical examples, a composite Representative Volume Element with elastic fibers and elasto-plastic matrix material is used as the microscopic model. The network is tested in a series of challenging scenarios and its performance is compared to that of a state-of-the-art Recurrent Neural Network (RNN). A remarkable outcome of the novel framework is the ability to naturally predict unloading/reloading behavior without ever seeing it during training, a stark contrast with popular but data-hungry models such as RNNs. Finally, the proposed network is applied to FE2 examples to assess its robustness for application in nonlinear finite element analysis. Subject Artificial Neural Networks (ANNs)Heterogeneous materialsMultiscalePath-dependency To reference this document use: http://resolver.tudelft.nl/uuid:3aed144e-3d70-4a10-977d-d9883de72daa DOI https://doi.org/10.1016/j.cma.2023.115934 ISSN 0045-7825 Source Computer Methods in Applied Mechanics and Engineering, 407 Part of collection Institutional Repository Document type journal article Rights © 2023 M. Alves Maia, I.B.C.M. Rocha, P. Kerfriden, F.P. van der Meer Files PDF 1_s2.0_S0045782523000579_main.pdf 4.38 MB Close viewer /islandora/object/uuid:3aed144e-3d70-4a10-977d-d9883de72daa/datastream/OBJ/view