{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,14]],"date-time":"2026-03-14T15:11:06Z","timestamp":1773501066535,"version":"3.50.1"},"reference-count":37,"publisher":"MDPI AG","issue":"12","license":[{"start":{"date-parts":[[2023,11,25]],"date-time":"2023-11-25T00:00:00Z","timestamp":1700870400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Algorithms"],"abstract":"<jats:p>Finite element (FE) simulations have been effective in simulating thermomechanical forming processes, yet challenges arise when applying them to new materials due to nonlinear behaviors. To address this, machine learning techniques and artificial neural networks play an increasingly vital role in developing complex models. This paper presents an innovative approach to parameter identification in flow laws, utilizing an artificial neural network that learns directly from test data and automatically generates a Fortran subroutine for the Abaqus standard or explicit FE codes. We investigate the impact of activation functions on prediction and computational efficiency by comparing Sigmoid, Tanh, ReLU, Swish, Softplus, and the less common Exponential function. Despite its infrequent use, the Exponential function demonstrates noteworthy performance and reduced computation times. Model validation involves comparing predictive capabilities with experimental data from compression tests, and numerical simulations confirm the numerical implementation in the Abaqus explicit FE code.<\/jats:p>","DOI":"10.3390\/a16120537","type":"journal-article","created":{"date-parts":[[2023,11,27]],"date-time":"2023-11-27T03:35:06Z","timestamp":1701056106000},"page":"537","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":12,"title":["Comparing Activation Functions in Machine Learning for Finite Element Simulations in Thermomechanical Forming"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7367-5453","authenticated-orcid":false,"given":"Olivier","family":"Pantal\u00e9","sequence":"first","affiliation":[{"name":"Laboratoire G\u00e9nie de Production, Institut National Polytechnique\/Ecole Nationale d\u2019Ing\u00e9nieurs de Tarbes, Universit\u00e9 de Toulouse, 47 Av d\u2019Azereix, F-65016 Tarbes, France"}]}],"member":"1968","published-online":{"date-parts":[[2023,11,25]]},"reference":[{"key":"ref_1","unstructured":"Abaqus (1989). Reference Manual, Hibbitt, Karlsson and Sorensen Inc."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"88","DOI":"10.1016\/j.msea.2007.11.119","article-title":"Modeling of flow stress of 42CrMo steel under hot compression","volume":"499","author":"Lin","year":"2009","journal-title":"Mater. Sci. Eng. A"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"125","DOI":"10.1016\/j.commatsci.2010.07.016","article-title":"A critical analysis of plastic flow behaviour in axisymmetric isothermal and Gleeble compression testing","volume":"50","author":"Bennett","year":"2010","journal-title":"Comput. Mater. Sci."},{"key":"ref_4","first-page":"81","article-title":"Thermo-mechanical simulation using gleeble system-advantages and limitations","volume":"58","author":"Kumar","year":"2016","journal-title":"J. Metall. Mater. Sci."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1039","DOI":"10.1016\/j.jmst.2018.12.026","article-title":"Refining constitutive relation by integration of finite element simulations and Gleeble experiments","volume":"35","author":"Yu","year":"2019","journal-title":"J. Mater. Sci. Technol."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"676","DOI":"10.1088\/0370-1301\/62\/11\/302","article-title":"An Investigation of the Mechanical Properties of Materials at very High Rates of Loading","volume":"62","author":"Kolsky","year":"1949","journal-title":"Proc. Phys. Soc. Sect. B"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"36","DOI":"10.1016\/S0749-6419(00)00097-8","article-title":"Unified Stress Update Algorithms for the Numerical Simulation of Large Deformation Elasto-Plastic and Elasto-Viscoplastic Processes","volume":"18","author":"Ponthot","year":"2002","journal-title":"Int. J. Plast."},{"key":"ref_8","unstructured":"Johnson, G.R., and Cook, W.H. (1983, January 19\u201321). A Constitutive Model and Data for Metals Subjected to Large Strains, High Strain Rates, and High Temperatures. Proceedings of the 7th International Symposium on Ballistics, The Hague, The Netherlands."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"1816","DOI":"10.1063\/1.338024","article-title":"Dislocation-mechanics-based constitutive relations for material dynamics calculations","volume":"61","author":"Zerilli","year":"1987","journal-title":"J. Appl. Phys."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1179\/095066069790138056","article-title":"Strength and structure under hot-working conditions","volume":"14","author":"Jonas","year":"1969","journal-title":"Metall. Rev."},{"key":"ref_11","unstructured":"Gao, C.Y. (2007). Computational Mechanics, Springer."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"308","DOI":"10.1051\/meca\/2018021","article-title":"An Efficient and Robust VUMAT Implementation of Elastoplastic Constitutive Laws in Abaqus\/Explicit Finite Element Code","volume":"19","author":"Ming","year":"2018","journal-title":"Mech. Ind."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"2200443","DOI":"10.1002\/srin.202200443","article-title":"A Modified Arrhenius-Type Constitutive Model and its Implementation by Means of the Safe Version of Newton\u2013Raphson Method","volume":"94","author":"Liang","year":"2022","journal-title":"Steel Res. Int."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Tize Mha, P., Dhondapure, P., Jahazi, M., Tongne, A., and Pantal\u00e9, O. (2023). Interpolation and extrapolation performance measurement of analytical and ANN-based flow laws for hot deformation behavior of medium carbon steel. Metals, 13.","DOI":"10.3390\/met13030633"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"103647","DOI":"10.1016\/j.finel.2021.103647","article-title":"Efficient implementation of non-linear flow law using neural network into the Abaqus Explicit FEM code","volume":"198","author":"Tongne","year":"2022","journal-title":"Finite Elem. Anal. Des."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Pantal\u00e9, O. (2023). Development and Implementation of an ANN Based Flow Law for Numerical Simulations of Thermo-Mechanical Processes at High Temperatures in FEM Software. Algorithms, 16.","DOI":"10.3390\/a16010056"},{"key":"ref_17","unstructured":"Minsky, M.L., and Papert, S. (1969). Perceptrons; An Introduction to Computational Geometry, MIT Press."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"359","DOI":"10.1016\/0893-6080(89)90020-8","article-title":"Multilayer Feedforward Networks Are Universal Approximators","volume":"2","author":"Hornik","year":"1989","journal-title":"Neural Net."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"103972","DOI":"10.1016\/j.jmps.2020.103972","article-title":"On the Potential of Recurrent Neural Networks for Modeling Path Dependent Plasticity","volume":"143","author":"Gorji","year":"2020","journal-title":"J. Mech. Phys. Solids"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"446","DOI":"10.1016\/j.measurement.2019.02.034","article-title":"The Sustainability of Neural Network Applications within Finite Element Analysis in Sheet Metal Forming: A Review","volume":"138","author":"Jamli","year":"2019","journal-title":"Measurement"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"752","DOI":"10.1016\/j.commatsci.2008.01.039","article-title":"Application of Neural Networks to Predict the Elevated Temperature Flow Behavior of a Low Alloy Steel","volume":"43","author":"Lin","year":"2008","journal-title":"Comput. Mater. Sci."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"102","DOI":"10.1016\/j.tws.2018.06.035","article-title":"Artificial Neural Networks and Intelligent Finite Elements in Non-Linear Structural Mechanics","volume":"131","author":"Stoffel","year":"2018","journal-title":"Thin-Walled Struct."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"85","DOI":"10.1016\/j.mechrescom.2019.01.004","article-title":"Neural Network Based Constitutive Modeling of Nonlinear Viscoplastic Structural Response","volume":"95","author":"Stoffel","year":"2019","journal-title":"Mech. Res. Commun."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"205","DOI":"10.1016\/j.ijplas.2019.05.001","article-title":"Application of Artificial Neural Networks in Micromechanics for Polycrystalline Metals","volume":"120","author":"Ali","year":"2019","journal-title":"Int. J. Plast."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"141259","DOI":"10.1016\/j.msea.2021.141259","article-title":"Artificial neural network modified constitutive descriptions for hot deformation and kinetic models for dynamic recrystallization of novel AZE311 and AZX311 alloys","volume":"816","author":"Wang","year":"2021","journal-title":"Mater. Sci. Eng. A"},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Cheng, P., Wang, D., Zhou, J., Zuo, S., and Zhang, P. (2022). Comparison of the Warm Deformation Constitutive Model of GH4169 Alloy Based on Neural Network and the Arrhenius Model. Metals, 12.","DOI":"10.3390\/met12091429"},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Churyumov, A.Y., and Kazakova, A.A. (2023). Prediction of True Stress at Hot Deformation of High Manganese Steel by Artificial Neural Network Modeling. Materials, 16.","DOI":"10.3390\/ma16031083"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"92","DOI":"10.1016\/j.neucom.2022.06.111","article-title":"Activation functions in deep learning: A comprehensive survey and benchmark","volume":"503","author":"Dubey","year":"2022","journal-title":"Neurocomputing"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"21","DOI":"10.1615\/JMachLearnModelComput.2023047367","article-title":"How important are activation functions in regression and classification? A survey, performance comparison, and future directions","volume":"4","author":"Jagtap","year":"2023","journal-title":"J. Mach. Learn. Model. Comput."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Mira, J., and Sandoval, F. (1995). From Natural to Artificial Neural Computation, Springer.","DOI":"10.1007\/3-540-59497-3"},{"key":"ref_31","unstructured":"Leen, T., Dietterich, T., and Tresp, V. (2000). Advances in Neural Information Processing Systems, MIT Press."},{"key":"ref_32","unstructured":"Ramachandran, P., Zoph, B., and Le, Q.V. (2018). Searching for Activation Functions. arXiv."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"160","DOI":"10.1016\/j.neunet.2021.04.011","article-title":"Neural network approximation: Three hidden layers are enough","volume":"141","author":"Shen","year":"2021","journal-title":"Neural Net."},{"key":"ref_34","unstructured":"Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., and Corrado, G.S. (2023, July 05). TensorFlow: Large-Scale Machine Learning on Heterogeneous Systems. 2015. Software. Available online: tensorflow.org."},{"key":"ref_35","unstructured":"Kingma, D.P., and Lei, J. (2014). Adam: A Method for Stochastic Optimization. arXiv."},{"key":"ref_36","unstructured":"Pantal\u00e9, O. (2023, November 15). Comparing Activation Functions in Machine Learning for Finite Element Simulations in Thermomechanical Forming: Software Source Files. Software Heritage. Available online: https:\/\/archive.softwareheritage.org\/swh:1:dir:b418ca8e27d05941c826b78a3d8a13b07989baf6."},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Koranne, S. (2011). Handbook of Open Source Tools, Springer.","DOI":"10.1007\/978-1-4419-7719-9"}],"container-title":["Algorithms"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1999-4893\/16\/12\/537\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T21:30:09Z","timestamp":1760131809000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1999-4893\/16\/12\/537"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,11,25]]},"references-count":37,"journal-issue":{"issue":"12","published-online":{"date-parts":[[2023,12]]}},"alternative-id":["a16120537"],"URL":"https:\/\/doi.org\/10.3390\/a16120537","relation":{},"ISSN":["1999-4893"],"issn-type":[{"value":"1999-4893","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,11,25]]}}}