Paper Titles
Explainable Machine Learning and 3D Visualization for Rotary Tube Bending: Expert Evaluation of a Web-Based Tool
p.43
Investigation of Code Optimization Strategies for Enhancing the Performance of Static Recrystallization Cellular Automata Models
p.59
Parameter Optimization and Data-Driven Soft-Sensor Framework for Torque Prediction in Bobbin-Tool Friction Stir Welding of AA2024
p.69
Long Short-Term Memory to Predict the Quality Parameters in the Hot Rolling Process
p.81
AI-Based Predictions of Forming Effects for Enhanced Crash Simulation
p.93
Residual Learning-Based Synthetic Data for Hybrid Metamodeling in the Stamping of an Automotive Door Panel
p.103
Robust and Feature-Aware Arbritary Lagrangian-Eulerian Method for Material Forming Applications
p.113
On-Line Prediction of Dry Zones during Composite Process through Digital Shadow Approach
p.123
Graph Neural Network for Draw-in Prediction in Sheet Metal Stamping
p.137

AI-Based Predictions of Forming Effects for Enhanced Crash Simulation

Article Preview
View Pdf By email

Abstract:

In the early concept phase of vehicle development, crash simulations must provide reliable statements on energy absorption and failure behavior, although detailed forming simulations and process data are typically not yet available. Manufacturing-induced material states, such as plastic pre-strains and local sheet thickness distributions, are therefore often neglected or approximated using simplified low-fidelity approaches with high uncertainty. This contribution introduces UmMatCraML, a data-driven medium-fidelity method for rapid initialization of typical crash shell meshes (element edge lengths 3–5 mm) with forming-induced field variables. Starting from the final component geometry, a purely geometric unfolding is performed to approximate the blank, from which a mesh-and coordinate-independent areal strain (Ar) is determined. A monotonicity-preserving gradient boosting regressor subsequently compensates for systematic deviations of this geometric surrogate model compared to high-fidelity deep drawing simulations, with four Hockett-Sherby parameters consistently parameterizing the material description. The training data are generated from approximately 5,000 LS-DYNA forming simulations and cover a broadly varied, physically consistent parameter space. In validation on a demonstrator part, UmMatCraML reduces the computation time for determining forming-induced component properties from about 60 min for High-Fidelity-Simualtion (HFS) or 10–15 min for Low-Fidelity-Simulation (LFS) to under 10 s, with simultaneously improved prediction quality. Demonstrations on components of a Toyota Yaris full-vehicle model show robust predictions even with trimming and perforations. Limitations arise from the model assumptions made (e.g., isotropic hardening, limited mapping of multi-stage process paths). Overall, UmMatCraML enables real-time, reproducible provision of manufacturing-induced field variables for concept crash simulations without explicit tool modeling.

You have full access to the following eBook
Advanced Numerical and AI Strategies for Material Forming Read eBook

Info:

Periodical:

Key Engineering Materials (Volume 1050)

Pages:

93-102

DOI:

https://doi.org/10.4028/p-kJ2Cti

Citation:

Cite this paper

Online since:

April 2026

Authors:

Ingolf Lepenies*

Keywords:

Deep Drawing, Forming Effects, Physics-Informed AI, Virtual Vehicle Development

Export:

RIS, BibTeX

Permissions:

Creative Commons CC BY 4.0

Share:

Citation:

* - Corresponding Author

References

[1] Erman Tekkaya, A.: State-of-the-art of simulation of sheet metal forming, Journal of Materials Processing Technology, Volume 103, Issue 1, 2000, Pages 14-22.

DOI: 10.1016/s0924-0136(00)00413-1

Google Scholar

[2] Gassler, L.; Haberthür, M.; Afrasiabi, M.; Bambach, M.: Graph Neural Network-based Generation of Stamped Bipolar Plate Geometries from Tooling Surfaces. Journal of Physics: Conference Series 3104 (2025) 012059. IOP Publishing.

DOI: 10.1088/1742-6596/3104/1/012059

Google Scholar

[3] Zhao, Y.; Chen, Q.; Li, H.; Zhou, H.; Attar, H. R.; Pfaff, T.: Recurrent U-Net- based Graph Neural Network (RUGNN) for accurate deformation predictions in sheet material forming. Advanced Engineering Informatics, Vol. 69, Part D, January 2026, Art.-Nr. 104021. ISSN 1474-0346.

DOI: 10.1016/j.aei.2025.104021

Google Scholar

[4] Hockett, J.; Sherby, O.: Large strain deformation of polycrystalline metals at low homologous temperatures. Journal of the Mechanics and Physics of Solids 23(2), 87-98 (1975).

DOI: 10.1016/0022-5096(75)90018-6

Google Scholar

[5] SCALE.sdm - Integrative Software Solution for Continuous Simulation Data and Process, Management. https://www.scale.eu/en/products/scale-sdm/ (2025).

Google Scholar

[6] Inal, K.; Lévesque, J., Worswick, M.: The 12th International Conference and Workshop on Numerical Simulation of 3D Sheet Metal Forming Processes (NUMISHEET 2022) (2022).

DOI: 10.1007/978-3-031-06212-4

Google Scholar

[7] Schwarzer, R.: Finite Element Model "Kirchhoff-Automotive-Demonstrator" (KADemonstrator) - Process Model for Deep Drawing. KIRCHHOFF Automotive Deutschland GmbH (2025).

Google Scholar

[8] Lepenies, I.; Krause, P.: Influence of Sheet Forming Using Machine Learning for Concept Crash Simulations. FAT Publication Series No. 395. Research Association for Automotive Engineering (FAT), Berlin 2025.

Google Scholar

[9] 2010 Toyota Yaris Finite Element Model. https://www.ccsa.gmu.edu/models/2010-toyota-yaris/.

Google Scholar

[10] Banabic, D.: Sheet Metal Forming Processes. Constitutive Modelling and Numerical Simulation.

Google Scholar

[11] Roll, K.:Including the effect of manufacturing in the design of products: crash properties. Proc. Int. Conf. Product Property Prediction, Dortmund, (2010).

Google Scholar

[12] Zhu, X.; Zhang, L.: Advance in Sheet Metal Forming - One-step Solution, Multi-Beads, Gravity Prebending, Auto Nets, and Local Compensation 12th International LS-DYNA Users Conference, 2012, 12.

Google Scholar

[13] Güner, A.; Soyarslan, C.; Brosius, A.; Tekkaya, A. Characterization of anisotropy of sheet metals employing inhomogeneous strain fields for Yld2000-2D yield function International Journal of Solids and Structures , 2012, 49, 3517 - 3527.

DOI: 10.1016/j.ijsolstr.2012.05.001

Google Scholar

[14] Nielsen, C. V.; Martins, P. A.: Metal Forming: Formability, Simulation, and Tool Design Academic Press, (2021).

Google Scholar