0636164 20260226134049.920250610235959.9 105007648898 001504431300001 10.1007/s00707-025-04385-8 18 s. E cav_un_epca*02560620001-59702368 (2025)4311-4328Springer Neural Network Cyclic Plastic Loading Parameter Estimation Non-gradient optimization cav_un_auth*0488850 Kovanda Martin UTIA-B Stochastická informatika Department of Stochastic Informatics SI SI CZ 40 K Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0283918 Marek René UT-L D 4 - Rázy a vlny v tělesech D 4 - Impact and Waves in Solids CZ 30 Ústav termomechaniky AV ČR, v. v. i. cav_un_auth*0101212 Tichavský Petr UTIA-B Stochastická informatika Department of Stochastic Informatics SI SI Department of Stochastic Informatics 30 Ústav teorie informace a automatizace AV ČR, v. v. i. http://library.utia.cas.cz/separaty/2025/SI/tichavsky-0636164.pdf https://link.springer.com/article/10.1007/s00707-025-04385-8 GA23-05338S GA ČR cav_un_auth*0452392 GA22-11101S GA ČR CZ cav_un_auth*0435406 EH23_020/0008501 GA MŠk CZ cav_un_auth*0477721 The increasing complexity of modern constitutive models of cyclic metal plasticity requires more efficient ways to achieve their optimal calibration. Traditional approaches, such as random search combined with Nelder-Mead optimization, are computationally expensive. In addition, they struggle with highly non-convex functions that have numerous local minima and complex behavior, making these methods highly sensitive to initial conditions. While numerical refinement is key, a better prediction for its initial point directly saves costs. In this work, we focus only on the uniaxial cyclic loading, as it is the dominant part of a general calibration process for such a model and can also utilize a closed-form solution, further speeding up the procedure. We propose a neural network framework with a loss function that combines the loss on both the predicted parameters and the generated stress responses. This network is then used to predict an initial point for Nelder-Mead optimization. Our method was also compared to the non-gradient Tensor Train Optimization method on both synthetic data and measured experiments. WOS JL 20000 20300 20301 2026 3 UT-L 10000 10200 10201 RVO:67985556 RVO:61388998 https://hdl.handle.net/11104/0367699 S A MECHANICS 2.5 0.9 6.9 0.00524 0.442 8999 93 0.598 47.89 3.04 2650 Q1 2.300 362 Q2 64 0.6 Q2 64 2025 105007648898 SCOPUS PUBMED 001504431300001 WOS 767 kB cav_un_epca*0256062 Acta Mechanica 236 8 2025 4311 4328 0001-5970 1619-6937 Springer