Skip to main content

Interpretable Machine Learning for Texture-Dependent Constitutive Models with Automatic Code Generation for Topological Optimization

Abstract

Genetic programming-based symbolic regression (GPSR) is a machine learning method which produces symbolic models that can be readily interpreted. This study utilized GPSR to derive uniaxial texture-based constitutive models for an additively manufactured alloy which were evaluated in post hoc analyses. Training data consisted of microscopy and mechanical testing data provided by the Air Force Research Laboratory (AFRL) which was supplemented using a viscoplastic model calibrated to the observed data. The validity of the calibrated crystal plasticity viscoplastic model is demonstrated as part of the 2019 AFRL Additive Manufacturing Modeling Challenge Series. Additionally, an expression evaluator was developed to integrate the constitutive models into the topology optimization software package Plato. A significant aspect of this paper is the presentation of these topics as components within a highly automated framework that allows efficient incorporation of microstructural characteristics into design activities. A topology optimization example was conducted using the GPSR results that constitutes application of the automated framework and post hoc analyses of the GPSR models demonstrate interpretability, suitability, and a probabilistic method to quantify domain bounds.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15
Fig. 16
Fig. 17
Fig. 18
Fig. 19

References

  1. 1.

    Mohd Yusuf S, Cutle S, Gao N (2019) Review: the impact of metal additive manufacturing on the aerospace industry. Metals 9(12):1286. https://doi.org/10.3390/met9121286

    CAS  Article  Google Scholar 

  2. 2.

    DebRoy T, We HL, Zuback JS, Mukherjee T, Elme JW, Milewski JO, Beese AM, Wilson-Heid A, De A, Zhang W (2018) Additive manufacturing of metallic components: process, structure and properties. Prog Mater Sci 92:112–224. https://doi.org/10.1016/j.pmatsci.2017.10.001

    CAS  Article  Google Scholar 

  3. 3.

    DebRoy T, Mukherjee T, Milewski JO, Elmer JW, Ribic B, Blecher JJ, Zhang W (2019) Scientific, technological and economic issues in metal printing and their solutions. Nature 18:1026–1032. https://doi.org/10.1038/s41563-019-0408-2

    CAS  Article  Google Scholar 

  4. 4.

    Koza JR (1994) Genetic programming as a means for programming computers by natural selection. Stat Comput 4:87–112. https://doi.org/10.1007/BF00175355

    Article  Google Scholar 

  5. 5.

    Schmidt M, Lipson H (2009) Distilling free-form natural laws from experimental data. Science 324(5923):81–85. https://doi.org/10.1126/science.1165893

    CAS  Article  Google Scholar 

  6. 6.

    Wang Y, Wagner N, Rondinelli J (2019) Symbolic regression in materials science. MRS Commun 9(3):793–805. https://doi.org/10.1557/mrc.2019.85

    CAS  Article  Google Scholar 

  7. 7.

    Air Force Research Laboratory (AFRL) (2020) Additive Manufacturing (AM) Modeling Challenge Series. https://materials-data-facility.github.io/MID3AS-AM-Challenge/. Accessed on May 18, 2021

  8. 8.

    Lebensohn RA, Tomé CN (1993) A self-consistent anisotropic approach for the simulation of plastic deformation and texture development of polycrystals: Application to zirconium alloys. Acta Metall Mater 41(9):2611–2624. https://doi.org/10.1016/0956-7151(93)90130-K

    CAS  Article  Google Scholar 

  9. 9.

    Groeber MA, Jackson MA (2014) DREAM.3D: a digital representation environment for the analysis of microstructure in 3D. Integr Mater Manuf Innov 3:56–72. https://doi.org/10.1186/2193-9772-3-5

    Article  Google Scholar 

  10. 10.

    Tomé CN, Lebensohn RA (2009) Manual for code visco-plastic self-consistent (VPSC). Version 7c. https://public.lanl.gov/lebenso/VPSC7c_manual.pdf. Accessed May 1 2020

  11. 11.

    Bomarito G, Townsend T, Hochhalter J, Adams E, Esham K, Vera D (2018) Bingo. https://github.com/nasa/bingo. Accessed on May 18, 2021

  12. 12.

    Plato Development Team (2021) Plato. Sandia National Laboratories. https://www.sandia.gov/plato3d/index.html. Accessed May 18, 2021

  13. 13.

    Salvatier J, Wiecki TV, Fonnesbeck C (2016) Probabilistic programming in Python using PyMC3. PeerJ Comput Sci 2:e55. https://doi.org/10.7717/peerj-cs.55

    CAS  Article  Google Scholar 

  14. 14.

    Roter F, Eisenlohr P, Hantcherli L, Tjahjanto DD, Bieler TR, Raabe D (2010) Overview of constitutive laws, kinematics, homogenization and multiscale methods in crystal plasticity finite-element modeling: theory, experiments, applications. Acta Mater 58(4):1152–1211. https://doi.org/10.1016/j.actamat.2009.10.058

    CAS  Article  Google Scholar 

  15. 15.

    Molinari A, Canova GR, Ahzi S (1987) A self consistent approach of the large deformation polycrystal viscoplasticity. Acta Metall 35(12):2983–2994. https://doi.org/10.1016/0001-6160(87)90297-5

    CAS  Article  Google Scholar 

  16. 16.

    Ripoll MR, Očenášek J (2009) Microstructure and texture evolution during the drawing of tungsten wires. Eng Fract Mech 76(10):1485–1499. https://doi.org/10.1016/j.engfracmech.2009.02.012

    Article  Google Scholar 

  17. 17.

    Maldar A, Wang L, Zhu G, Zeng X (2020) Investigation of the alloying effect on deformation behavior in Mg by Visco-Plastic Self-Consistent modeling. J Magn Alloys 8(1):210–218. https://doi.org/10.1016/j.jma.2019.07.009

    CAS  Article  Google Scholar 

  18. 18.

    Zhang K, Holmedal B, Hopperstad OS, Dumoulin S, Gawad S, Van Bael A, Van Houtte P (2015) Multi-level modelling of mechanical anisotropy of commercial pure aluminium plate: Crystal plasticity models, advanced yield functions and parameter identification. Int J Plast 66:3–30. https://doi.org/10.1016/j.ijplas.2014.02.003

    CAS  Article  Google Scholar 

  19. 19.

    Kotha S, Ozturk D, Ghosh S (2019) Parametrically homogenized constitutive models (PHCMs) from micromechanical crystal plasticity FE simulations, part I: Sensitivity analysis and parameter identification for Titanium alloys. Int J Plast 120:296–319. https://doi.org/10.1016/j.ijplas.2019.05.008

    CAS  Article  Google Scholar 

  20. 20.

    Weber G, Pinz M, Gosh S (2020) Machine learning-aided parametrically homogenized crystal plasticity model (PHCPM) for single crystal Ni-based superalloys. JOM 72:4404–4419. https://doi.org/10.1007/s11837-020-04344-9

    Article  Google Scholar 

  21. 21.

    Versino D, Tonda A, Bronkhorst CA (2017) Data driven modeling of plastic deformation. Comput Methods Appl Mech Eng 318:981–1004. https://doi.org/10.1016/j.cma.2017.02.016

    Article  Google Scholar 

  22. 22.

    Bomarito G, Townsend TS, Stewart KM, Esham KV, Emery JM, Hochhalter JD (2021) Development of interpretable, data-driven plasticity models with symbolic regression. Comput Struct 252:0045–7949. https://doi.org/10.1016/j.compstruc.2021.106557

    Article  Google Scholar 

  23. 23.

    Baron JL, Esling C, Feron JL, Gex D, Glimois JL, Guillen G, Humbert M, Lemoine P, Lepape J, Mardon JP, Thil A, Uny G (1990) Interlaboratories tests of textures of zircaloy-4 tubes. 1. Pole figure measurements and calculation of kearns coefficients. Textures Microstruct. 12:125–140

    Google Scholar 

  24. 24.

    Voce E (1948) The relationship between stress and strain for homogeneous deformation. J Instit Metals. 74:537–562

    CAS  Google Scholar 

  25. 25.

    Knuth DE (1997) The art of computer programming. Volume 1 fundamental algorithms (3 ed.). Addison Wesley. Hre: Sect.2.3.1 “Traversing Binary Trees”

  26. 26.

    Edwards HC, Trott CR, Sunderland D (2014) Enabling manycore performance portability through polymorphic memory access patterns. J Parallel Distrib Comput 74(12):3202–3216. https://doi.org/10.1016/j.jpdc.2014.07.003

    Article  Google Scholar 

Download references

Acknowledgements

The authors acknowledge Sandia National Laboratories, a multi-mission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International, Inc., for the US Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525. The support and resources from the Center for High Performance Computing at the University of Utah are gratefully acknowledged. The authors also recognize the developers of Bingo, Dr. Geoff Bomarito and Dr. Patrick Lesser, for their contributions and support. The efforts which contributed to data collection and organization of the challenge are gratefully appreciated. Professor A. Rollett acknowledges support from US Department of Energy, Office of Science, Basic Energy Sciences, under Award # DE-SC0019096.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Karl Garbrecht.

Ethics declarations

Conflict of Interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

Appendix 1: Genetic Programming-Based Symbolic Regression Hyper-Parameter Summary

Appendix 1: Genetic Programming-Based Symbolic Regression Hyper-Parameter Summary

See Table 2.

Table 2 Islands are computing cores used during each run

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Garbrecht, K., Aguilo, M., Sanderson, A. et al. Interpretable Machine Learning for Texture-Dependent Constitutive Models with Automatic Code Generation for Topological Optimization. Integr Mater Manuf Innov 10, 373–392 (2021). https://doi.org/10.1007/s40192-021-00231-6

Download citation

Keywords

  • Symbolic regression
  • Constitutive model
  • Texture
  • Topology optimization
  • Viscoplastic
  • Parameter homogenization