Comparative study of artificial neural network and physics-informed neural network application in sheet metal forming

Comparative study of artificial neural network and physics-informed neural network application in sheet metal forming


download PDF

Abstract. Accurate prediction of the resultant geometry in sheet metal forming simulation is necessary to achieve zero-defect production. To quantify the effect of process parameters on the final geometry, numerical methods are used to simulate the process outputs for a given set of process variables. Finite element methods are employed in process optimization and design exploration. However, these computationally expensive models are unhelpful for process control applications. Surrogate models allowing fast prediction of resultant geometry or stress distribution can be plausible solutions. In the current study, we propose a sequential surrogate model to fit the stress field as a function of the process variable and the initial spatial coordinates. The framework is composed of two surrogate models. First, an artificial neural network (ANN) evaluates the displacement and the strain. Then, a second surrogate is employed to fit the stress using input strain and displacement. Here, ANN and physics-informed neural networks (PINN) are compared concerning prediction accuracy for the second surrogate model. The PINN is enhanced with the equilibrium equations. The developed method is demonstrated using a v-bending process. The results show that both surrogate models return good approximations, with ANN showing slightly better results.

Surrogate Modeling, Physics-Informed Neural Networks, Metal-Forming

Published online 4/24/2024, 11 pages
Copyright © 2024 by the author(s)
Published under license by Materials Research Forum LLC., Millersville PA, USA

Citation: MUNZONE Francesco, HAZRATI Javad, HAKVOORT Wouter, VAN DEN BOOGAARD Ton, Comparative study of artificial neural network and physics-informed neural network application in sheet metal forming, Materials Research Proceedings, Vol. 41, pp 2278-2288, 2024


The article was published as article 251 of the book Material Forming

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 license. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.

[1] I. El Mrabti, A. El Hakimi, A. Touache and C. Abderrahim, “A comparative study of surrogate models for predicting process failures during the sheet metal forming process of advanced high-strength steel,” The International Journal of Advanced Manufacturing Technology, vol. 121, no. 1, pp. 199-214, 2022.
[2] F. Abbassi, T. Belhadj, S. Mistou and A. Zghal, “Parameter identification of a mechanical ductile damage using Artificial Neural Networks in sheet metal forming,” Material and Design, vol. 45, pp. 605-615, 2013.
[3] J. Ghaisari, H. Jannesari and M. Vatani, “Artificial neural network predictors for mechanical properties of cold rolling products,” Advance in Engineering Software, vol. 45, pp. 91-99, 2012.
[4] M. B. Gorji, M. Mozaffar, J. H. Heidenreich, J. Cao and D. Mohr, “On the potential of recurrent neural networks for modeling path dependent plasticity,” Journal of the Mechanics and Physics of Solids, vol. 143, p. 103972, 2020.
[5] J. G. Hoffer, B. C. Geiger, P. Ofner and R. Kern, “Mesh-Free Surrogate Models for Structural Mechanic FEM Simulation: A Comparative Study of Approaches,” Applied Sciences, vol. 11, no. 20, p. 9411, 2021.
[6] E. Haghighat, M. Raissi, A. Moure, H. Gomez and R. Juanes, “A physics-informed deep learning framework for inversion and surrogate modeling in solid mechanics,” Computer Methods in Applied Mechanics and Engineering, vol. 379, 2021.
[7] S. Niu, E. Zhang, Y. Bazilevs and V. Srivastava, “Modeling finite-strain plasticity using physics-informed neural network and assessment of the network performance,” J. Mech. Phys. Solids, vol. 172, p. 105177, 2023.
[8] B. de Gooijer, “Surrogate modelling and robust optimization of multi-stage metal forming processes,” 2022.
[9] H. Liu, J. Cai and Y. S. Ong, “A Survey of Adaptive Sampling for Global Metamodeling in Support of Simulation-based Complex Engineering Design,” Structural and Multidisciplinary Optimization, vol. 57, pp. 393-416, 2018.
[10] K. P. Murphy, Probabilistic Machine Learning: An introduction, MIT Press, 2022.
[11] M. Raissi, P. Perdikaris and G. E. Karniada, “Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations,” Journal of Computational Physics, vol. 378, pp. 686-707, 2019.
[12] S. Razavi, B. . A. Tolson and D. H. Burn, “Review of surrogate modeling in water resources,” Water Resources Research , 2012.
[13] K. Hornik, “Multilayer Feedforward Networks are Universal Approximators,” Neural Networks, vol. 2, pp. 359-366, 1989.
[14] R. Yondo, E. Andrés and E. Valero, “A review on design of experiments and surrogate models in aircraft real-time and many-query aerodynamic analyses,” Elsevier, 2020.
[15] A. J. Forrester and A. J. Kean, “Recent advances in surrogate-based optimization,” Progress in Aereospace Sciences, vol. 45, no. 1-3, pp. 50-79, 2009.
[16] R. Jin, W. Chen and T. W. Simpson, “Comparative studies of metamodeling techniques under multiple modeling criteria,” Structural Multidisciplinary Optimization, 2001.
[17] J. H. Wiebenga, A. H. van den Boogaard and G. Klaseboer, “Sequential robust optimization of a V-bending process using numerical simulations,” Struct Multidisc Optim, vol. 46, pp. 137-153, 2012.