Revisiting Preparation of Phase Space for Learning Path-Dependent Behavior via Deep Neural Networks
Publication: Journal of Engineering Mechanics
Volume 148, Issue 12
Abstract
This technical note investigates the preparation of a phase space of two simple constitutive laws, von Mises (J2) and Drucker-Prager (DP) models, for deep neural networks with special attention to elastic and plastic data points. The phase space is referred to as a database describing the material behavior via data points instead of mathematical formulation. To examine the effect of the proportions of elastic and plastic data points on the prediction quality, two sets (for J2 and DP models) including three equal phase spaces with different proportions of elastic and plastic data points are considered. To make a fair comparison, the only difference in phase spaces is the proportion of elastic and plastic data points in each set. We further study two different data types for constitutive behaviors, the tensorial space and the invariant space (volumetric-deviatoric space). The Nash-Sutcliffe error (NSE) is calculated for an unbiased comparison of prediction with different phase spaces in both tensorial and volumetric-deviatoric spaces. The results reveal that the distribution of elastic and plastic points may affect the accuracy of prediction for simple constitutive laws (isotropic, elasto-perfect plastic) via artificial neural networks; when an equal portion of elastic and plastic data points are used, the more robust prediction is achieved in this study.
Get full access to this article
View all available purchase options and get full access to this article.
Data Availability Statement
All data, models, or code that support the findings of this study are available from the corresponding author upon reasonable request.
Acknowledgments
This research was supported by Natural Sciences and Engineering Research Council of Canada (Discovery Grant, RGPIN-2019-06471). The support is gratefully acknowledged. The authors thank the associate editor and two anonymous reviewers for their helpful suggestions and comments.
References
Abadi, M., et al. 2015. “TensorFlow: Large-scale machine learning on heterogeneous systems.” Accessed May 31, 2022. https://www.tensorflow.org/.
Capuano, G., and J. J. Rimoli. 2019. “Smart finite elements: A novel machine learning application.” Comput. Methods Appl. Mech. Eng. 345 (1): 363–381. https://doi.org/10.1016/j.cma.2018.10.046.
Chollet, F. et al. 2015. “Keras.” Accessed May 31, 2022. https://github.com/fchollet/keras.
Ge, W., and V. L. Tagarielli. 2021. “A computational framework to establish data-driven constitutive models for time-or path-dependent heterogeneous solids.” Sci. Rep. 11 (1): 1–18. https://doi.org/10.1038/s41598-021-94957-0.
Ghaboussi, J., J. H. Garrett, and X. Wu. 1991. “Knowledge-based modeling of material behavior with neural networks.” J. Eng. Mech. 117 (1): 132–153. https://doi.org/10.1061/(ASCE)0733-9399(1991)117:1(132).
Ghavamian, F., and A. Simone. 2019. “Accelerating multiscale finite element simulations of history-dependent materials using a recurrent neural network.” Comput. Methods Appl. Mech. Eng. 357 (8): 112594. https://doi.org/10.1016/j.cma.2019.112594.
Gorji, M. B., and D. Mohr. 2019. “A basic neural network model describing the plasticity of sheet metal.” In Proc., 13th Int. Conf. on Numerical Methods in Industrial Forming Processes, IDDRG2019. Pittsburgh, PA: The Minerals, Metals & Materials Society.
Haghighi, E. M., and S. Na. 2021. “A multifeatured data-driven homogenization for heterogeneous elastic solids.” Appl. Sci. 11 (19): 9208. https://doi.org/10.3390/app11199208.
Haghighi, E. M., and S. Na. 2022. “A single long short-term memory network for enhancing the prediction of path-dependent plasticity with material heterogeneity and anisotropy.” Accessed August 31, 2022. https://arxiv.org/abs/2204.01466v2.
Hochreiter, S., and J. Schmidhuber. 1997. “Long short-term memory.” Neural Comput. 9 (8): 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735.
Homel, M., C. Sherman, and J. Morris. 2019. Machine learning for constitutive modeling on a graphics processing unit. Livermore, CA: Lawrence Livermore National Lab.
Im, S., J. Lee, and M. Cho. 2021. “Surrogate modeling of elasto-plastic problems via long short-term memory neural networks and proper orthogonal decomposition.” Comput. Methods Appl. Mech. Eng. 385 (5): 114030. https://doi.org/10.1016/j.cma.2021.114030.
Javadinasr, M., A. Mohammadian, and A. Parsa. 2021. “A deep-learning based optimization approach to address stop-skipping strategy in urban rail transit lines.” In Proc., Int. Conf. on Transportation and Development 2022, 165–176. Reston, VA: ASCE. https://ascelibrary.org/doi/abs/10.1061/9780784484319.016.
Logarzo, H. J., G. Capuano, and J. J. Rimoli. 2021. “Smart constitutive laws: Inelastic homogenization through machine learning.” Comput. Methods Appl. Mech. Eng. 373 (5): 113482. https://doi.org/10.1016/j.cma.2020.113482.
Ma, G., S. Guan, Q. Wang, Y. Feng, and W. Zhou. 2022. “A predictive deep learning framework for path-dependent mechanical behavior of granular materials.” Acta Geotech. 2022 (1): 1–16. https://doi.org/10.1007/s11440-021-01419-y.
Mozaffar, M., R. Bostanabad, W. Chen, K. Ehmann, J. Cao, and M. Bessa. 2019. “Deep learning predicts path-dependent plasticity.” Proc. Nat. Acad. Sci. 116 (52): 26414–26420. https://doi.org/10.1073/pnas.1911815116.
Nash, J. E., and J. V. Sutcliffe. 1970. “River flow forecasting through conceptual models part i–a discussion of principles.” J. Hydrol. 10 (3): 282–290. https://doi.org/10.1016/0022-1694(70)90255-6.
Palau, T., A. Kuhn, S. Nogales, H. Böhm, A. Rauh, and M. B. Ag. 2012. “A neural network based elasto-plasticity material model.” In CD-ROM Proc., 6th European Congress on Computational Methods in Applied Sciences and Engineering, 10. Vienna, Austria: TU Wien.
Smyrniou, E. 2018. “Soil constitutive modelling using neural networks.” Master’s thesis, Dept. of Geo-Engineering, Delft Univ. of Technology.
Information & Authors
Information
Published In
Journal of Engineering Mechanics
Volume 148 • Issue 12 • December 2022
Copyright
© 2022 American Society of Civil Engineers.
History
Received: Mar 2, 2022
Accepted: Aug 7, 2022
Published online: Sep 30, 2022
Published in print: Dec 1, 2022
Discussion open until: Feb 28, 2023
ASCE Technical Topics:
- Artificial intelligence and machine learning
- Comparative studies
- Computer programming
- Computing in civil engineering
- Constitutive relations
- Continuum mechanics
- Databases
- Deformation (mechanics)
- Elastic analysis
- Elastoplasticity
- Engineering fundamentals
- Engineering mechanics
- Information Technology (IT)
- Material mechanics
- Material properties
- Materials engineering
- Mathematical models
- Mathematics
- Methodology (by type)
- Models (by type)
- Neural networks
- Research methods (by type)
- Solid mechanics
- Structural analysis
- Structural engineering
- Structural mechanics
Authors
Metrics & Citations
Metrics
Citations
Download citation
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.