nach oben

Integrating Materials and Manufacturing Innovation

Erschienen in:

14.09.2022 | Technical Article

Application of a Chained-ANN for Learning the Process–Structure Mapping in Mg₂Si_xSn_1−x Spinodal Decomposition

verfasst von: Grayson H. Harrington, Conlain Kelly, Vahid Attari, Raymundo Arroyave, Surya R. Kalidindi

Erschienen in: Integrating Materials and Manufacturing Innovation | Ausgabe 3/2022

Einloggen, um Zugang zu erhalten

Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.

search-config

KI-gestützte Suche

Aus

Abstract

This work establishes a reliable and accurate materials process–structure (PS) surrogate model that maps an 18-dimensional process parameter input domain to a high-dimensional space of single- and dual-phase microstructures. This was accomplished by employing the Materials Knowledge Systems (MKS) framework (includes microstructure quantification via two-point statistics and dimensionality reduction using principal components analysis) for the feature engineering of the microstructures, and subsequently constructing a chained-artificial neural network (ANN) to learn the complex nonlinear mappings between the high-dimensional input domain and the MKS-derived low-dimensional representation of the corresponding microstructure space (includes both homogeneous and heterogeneous microstructures). The benefits of this workflow are demonstrated on a collection of ~ 10,000 final microstructures obtained from chemo-mechanical spinodal decomposition phase-field simulations in the Mg₂Si_xSn_1-x material system. Specifically, it is shown that the complex phase-field process–structure relationships for the selected case study can be captured in a robust model with only 742 fittable parameters.

Vorheriger Artikel Image Segmentation for Defect Analysis in Laser Powder Bed Fusion: Deep Data Mining of X-Ray Photography from Recent Literature

Nächster Artikel Microstructure Characterization and Reconstruction in Python: MCRpy

This usually incurs a one-time high computational cost.

Surrogates can also be formulated to predict microstructure evolution (i.e., predict the microstructure at the next time step based on known microstructures at previous time steps and known processing conditions) [8, 11, 17, 30, 31].

The 15,000^th time step corresponded to fully decomposed microstructures for most of the simulations. However, the phase separation was sluggish in a few cases and produced three-phase microstructures. As noted later, a small number of three-phase microstructures found in the extracted dataset were excluded from the study. Therefore, the microstructures obtained at the 15,000th time step served as the final microstructures for our study.

Voorhees P, Spanos G (2015) Modeling across scales: a roadmapping study for connecting materials models and simulations across length and time scales. TMS, Warrendale, PA, p 14

Materials Genome Initiative for Global Competitiveness, 2011. www.ostp.gov/nstc

Cummings P, Glotzer S. (2010) Inventing a new America through discovery and innovation in science, engineering and medicine: a vision for research and development in simulation-based engineering and science in the next decade. https://www.nsf.gov/mps/ResearchDirectionsWorkshop2010/RWD-color-FINAL-usletter_2010-07-16.pdf (accessed December 26, 2021)

Glotzer S, Kim S, Cummings P, Deshmukh A, Head M, Karniadakis G, Petzold L, Sagui C, Shinozuka M, (2009) WTEC Panel report on international assessment of research and development in simulation-based engineering and science. World Technology Evaluation Center, Inc., Baltimore, MD

National Research Council (2008) Integrated computational materials engineering: a transformational discipline for improved competitiveness and national security. National Academies Press, Washington, DC

Kalidindi SR (2015) Hierarchical materials informatics: novel analytics for materials data, Elsevier

Fullwood DT, Niezgoda SR, Adams BL, Kalidindi SR (2010) Microstructure sensitive design for performance optimization. Prog Mater Sci 55:477–562. https://doi.org/10.1016/J.PMATSCI.2009.08.002CrossRef

Brough DB, Wheeler D, Warren JA, Kalidindi SR (2017) Microstructure-based knowledge systems for capturing process-structure evolution linkages. Curr Opin Solid State Mater Sci 21:129–140. https://doi.org/10.1016/J.COSSMS.2016.05.002CrossRef

Sarkar D, Reddy BS, Basu B (2018) Implementing statistical modeling approach towards development of ultrafine grained bioceramics: case of ZrO₂-toughened Al₂O₃. J Am Ceram Soc 101:1333–1343CrossRef

10.

Tapia G, Johnson L, Franco B, Karayagiz K, Ma J, Arroyave R, Karaman I, Elwany A (2017) Bayesian calibration and uncertainty quantification for a physics-based precipitation model of nickel–titanium shape-memory alloys. J Manufact Sci Eng 139:071002CrossRef

11.

Yabansu YC, Steinmetz P, Hötzer J, Kalidindi SR, Nestler B (2017) Extraction of reduced-order process-structure linkages from phase-field simulations. Acta Mater 124:182–194. https://doi.org/10.1016/j.actamat.2016.10.071CrossRef

12.

Cecen A, Dai H, Yabansu YC, Kalidindi SR, Song L (2018) Material structure-property linkages using three-dimensional convolutional neural networks. Acta Mater 146:76–84CrossRef

13.

Popova E, Rodgers TM, Gong X, Cecen A, Madison JD, Kalidindi SR (2017) Process-structure linkages using a data science approach: application to simulated additive manufacturing data, integrating materials and manufacturing. Innovation 6:54–68. https://doi.org/10.1007/s40192-017-0088-1CrossRef

14.

Hassinger I, Li X, Zhao H, Xu H, Huang Y, Prasad A, Schadler L, Chen W, Catherine Brinson L (2016) Toward the development of a quantitative tool for predicting dispersion of nanocomposites under non-equilibrium processing conditions. J Mater Sci 51:4238–4249. https://doi.org/10.1007/s10853-015-9698-1CrossRef

15.

Gorgannejad S, Reisi Gahrooei M, Paynabar K, Neu RW (2019) Quantitative prediction of the aged state of Ni-base superalloys using PCA and tensor regression. Acta Materialia. 165:259–269. https://doi.org/10.1016/j.actamat.2018.11.047CrossRef

16.

Brough DB, Wheeler D, Kalidindi SR (2017) Materials knowledge systems in python—a data science framework for accelerated development of hierarchical materials. Integr Mater Manuf Innov 6:36–53CrossRef

17.

Hashemi S, Kalidindi SR (2021) A machine learning framework for the temporal evolution of microstructure during static recrystallization of polycrystalline materials simulated by cellular automaton. Comput Mater Sci 188:110132. https://doi.org/10.1016/j.commatsci.2020.110132CrossRef

18.

Hashemi S, Ganapathysubramanian B, Casey S, Su J, Kalidindi SR (2021) Feature engineering for microstructure-property mapping in organic photovoltaics. arXiv Preprint arXiv:2111.01897

19.

Kaundinya PR, Choudhary K, Kalidindi SR (2022) Prediction of the electron density of states for crystalline compounds with Atomistic Line Graph Neural Networks (ALIGNN), arXiv Preprint arXiv:2201.08348

20.

Pfeifer S, Wodo O, Ganapathysubramanian B (2018) An optimization approach to identify processing pathways for achieving tailored thin film morphologies. Comput Mater Sci 143:486–496CrossRef

21.

Timmermann S, Starostin V, Girelli A, Ragulskaya A, Rahmann H, Reiser M, Begam N, Randolph L, Sprung M, Westermeier F (2022) Automated matching of two-time X-ray photon correlation maps from phase-separating proteins with Cahn–Hilliard-type simulations using auto-encoder networks. J Appl Crystallogr. 55:751–757CrossRef

22.

Teichert GH, Natarajan AR, der Ven A, Garikipati K (2019) Machine learning materials physics: Integrable deep neural networks enable scale bridging by learning free energy functions. Comput Methods Appl Mech Eng 353:201–216CrossRef

23.

Teichert GH, Garikipati K (2019) Machine learning materials physics: Surrogate optimization and multi-fidelity algorithms predict precipitate morphology in an alternative to phase field dynamics. Comput Methods Appl Mech Eng 344:666–693CrossRef

24.

Zhang X, Garikipati K (2020) Machine learning materials physics: Multi-resolution neural networks learn the free energy and nonlinear elastic response of evolving microstructures. Comput Methods Appl Mech Eng 372:113362CrossRef

25.

Badawy MF, Msekh MA, Hamdia KM, Steiner MK, Lahmer T, Rabczuk T (2017) Hybrid nonlinear surrogate models for fracture behavior of polymeric nanocomposites. Probab Eng Mech 50:64–75CrossRef

26.

Kunselman C, Attari V, McClenny L, Braga-Neto U, Arroyave R (2020) Semi-supervised learning approaches to class assignment in ambiguous microstructures. Acta Mater 188:49–62CrossRef

27.

Kunselman C, Sheikh S, Mikkelsen M, Attari V, Arróyave R (2022) Microstructure classification in the unsupervised context. Acta Mater 223:117434CrossRef

28.

Farizhandi AA, Mamivand M (2022) Processing time, temperature, and initial chemical composition prediction from materials microstructure by deep network for multiple inputs and fused data. Mater Des 219:110799. https://doi.org/10.1016/j.matdes.2022.110799CrossRef

29.

Herman E, Stewart JA, Dingreville R (2020) A data-driven surrogate model to rapidly predict microstructure morphology during physical vapor deposition. Appl Math Model 88:589–603. https://doi.org/10.1016/J.APM.2020.06.046CrossRef

30.

Hu C, Martin S, Dingreville R (2022) Accelerating phase-field predictions via recurrent neural networks learning the microstructure evolution in latent space. Comput Methods Appl Mech Eng 397:115128CrossRef

31.

de Oca M, Zapiain D, Stewart JA, Dingreville R (2021) Accelerating phase-field-based microstructure evolution predictions via surrogate models trained by machine learning methods. Npj Computat Mater 7:4566. https://doi.org/10.1038/s41524-020-00471-8CrossRef

32.

Attari V, Sauceda DJ, Open Phase-field Microstructure Database (OPMD), (2019). http://microstructures.net/.

33.

Attari V, Honarmandi P, Duong T, Sauceda DJ, Allaire D, Arroyave R (2020) Uncertainty propagation in a multiscale CALPHAD-Reinforced elastochemical phase-field model. Acta Mater 183:452–470. https://doi.org/10.1016/j.actamat.2019.11.031CrossRef

34.

McClenny L, Haile M, Attari V, Sadler B, Braga-Neto U, Arroyave R (2020) Deep multimodal transfer-learned regression in data-poor domains, arXiv Preprint arXiv:2006.09310

35.

Adams BL, Gao X, Kalidindi SR (2005) Finite approximations to the second-order properties closure in single phase polycrystals. Acta Mater 53:3563–3577. https://doi.org/10.1016/J.ACTAMAT.2005.03.052CrossRef

36.

Niezgoda SR, Kanjarla AK, Kalidindi SR (2013) Novel microstructure quantification framework for databasing, visualization, and analysis of microstructure data. Integr Mater Manufact Innov 2:54–80. https://doi.org/10.1186/2193-9772-2-3CrossRef

37.

Niezgoda SR, Fullwood DT, Kalidindi SR (2008) Delineation of the space of 2-Point correlations in a composite material system. Acta Mater 56:5285–5292. https://doi.org/10.1016/J.ACTAMAT.2008.07.005CrossRef

38.

Generale AP, Kalidindi SR (2021) Reduced-order models for microstructure-sensitive effective thermal conductivity of woven ceramic matrix composites with residual porosity. Compos Struct 274:114399CrossRef

39.

Fullwood DT, Niezgoda SR, Kalidindi SR (2008) Microstructure reconstructions from 2-point statistics using phase-recovery algorithms. Acta Mater 56:942–948. https://doi.org/10.1016/j.actamat.2007.10.044CrossRef

40.

Robertson AE, Kalidindi SR (2022) Efficient generation of anisotropic N-field microstructures from 2-point statistics using multi-output Gaussian random fields. Acta Materialia. 232:117927CrossRef

41.

Izadi H, Baniassadi M, Hasanabadi A, Mehrgini B, Memarian H, Soltanian-Zadeh H, Abrinia K (2017) Application of full set of two point correlation functions from a pair of 2D cut sections for 3D porous media reconstruction. J Petrol Sci Eng 149:789–800CrossRef

42.

Hasanabadi A, Baniassadi M, Abrinia K, Safdari M, Garmestani H (2016) 3D microstructural reconstruction of heterogeneous materials from 2D cross sections: a modified phase-recovery algorithm. Comput Mater Sci 111:107–115. https://doi.org/10.1016/j.commatsci.2015.09.015CrossRef

43.

Dubey SR, Singh SK, Chaudhuri BB (2021) A comprehensive survey and performance analysis of activation functions in deep learning, arXiv Preprint arXiv:2109.14545

44.

Hecht-Nielsen, R. (1992) Theory of the backpropagation neural network, In: Neural networks for perception, Elsevier, pp. 65–93.

45.

Yi S, Attari V, Jeong M, Jian J, Xue S, Wang H, Arroyave R, Yu C (2018) Strain-induced suppression of the miscibility gap in nanostructured Mg 2 Si–Mg 2 Sn solid solutions. J Mater Chem A 6:17559–17570CrossRef

46.

Wang S, Mingo N (2009) Improved thermoelectric properties of Mg₂Si_xGe_ySn_1−x−y nanoparticle-in-alloy materials. Appl Phys Lett. https://doi.org/10.1063/1.3139785CrossRef

47.

Zhang Q, He J, Zhu TJ, Zhang SN, Zhao XB, Tritt TM (2008) High figures of merit and natural nanostructures in Mg₂Si_0.4Sn_0.6 based thermoelectric materials. Appl Phys Lett 93:25454. https://doi.org/10.1063/1.2981516CrossRef

48.

Otsu N (1979) A threshold selection method from gray-level histograms. IEEE Trans Syst Man Cybern 9:62–66CrossRef

49.

Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, Prettenhofer P, Weiss R, Dubourg V (2011) Scikit-learn: Machine learning in Python. J Mach Learn Res 12:2825–2830

50.

Paszke A, Gross S, Massa F, Lerer A, Bradbury J, Chanan G, Killeen T, Lin Z, Gimelshein N, Antiga L, (2019) Pytorch: An imperative style, high-performance deep learning library. Adv Neural Inf Process Syst. 32

51.

LeCun YA, Bottou L, Orr GB, Müller K-R (2012) Efficient backprop. Neural networks: tricks of the trade. Springer, USA, pp 9–48CrossRef

52.

Wang Q, Ma Y, Zhao K, Tian Y (2022) A comprehensive survey of loss functions in machine learning, Annals of Data. Science 9:187–212

53.

Loshchilov I, Hutter F (2018) Fixing weight decay regularization in adam

54.

Smith LN, Topin N. (2019) Super-convergence: very fast training of neural networks using large learning rates, In: Artificial intelligence and machine learning for multi-domain operations applications, pp. 369–386.

55.

Ioffe S, Szegedy C, (2015) Batch normalization: Accelerating deep network training by reducing internal covariate shift, In: International conference on machine learning, pp. 448–456

56.

Robertson AE, Kalidindi SR (2021) Digital representation and quantification of discrete dislocation structures. JOM 73:2143–2158CrossRef

57.

Harrington G, Kelly C. Chained-ANN-Phase-field-Surrogate, (2022). https://github.com/grayson-harrington/Chained-ANN-Phase-field-Surrogate

Titel: Application of a Chained-ANN for Learning the Process–Structure Mapping in Mg2SixSn1−x Spinodal Decomposition
verfasst von: Grayson H. Harrington
Conlain Kelly
Vahid Attari
Raymundo Arroyave
Surya R. Kalidindi
Publikationsdatum: 14.09.2022
Verlag: Springer International Publishing
Erschienen in: Integrating Materials and Manufacturing Innovation / Ausgabe 3/2022
Print ISSN: 2193-9764
Elektronische ISSN: 2193-9772
DOI: https://doi.org/10.1007/s40192-022-00274-3

Premium Partner

Marktübersichten

Die im Laufe eines Jahres in der „adhäsion“ veröffentlichten Marktübersichten helfen Anwendern verschiedenster Branchen, sich einen gezielten Überblick über Lieferantenangebote zu verschaffen.

Zur Marktübersicht

Springer Professional

Abstract

Bitte loggen Sie sich ein, um Zugang zu Ihrer Lizenz zu erhalten.

Weitere Artikel der Ausgabe 3/2022

Random Generation of Lattice Structures with Short-Range Order

On the Prediction of Uniaxial Tensile Behavior Beyond the Yield Point of Wrought and Additively Manufactured Ti-6Al-4V

Microstructure Characterization and Reconstruction in Python: MCRpy

Digital Protocols for Statistical Quantification of Microstructures From Microscopy Images of Polycrystalline Nickel-Based Superalloys

Coupled Process Modeling of Flow and Transport Phenomena in LCM Processing

Feature Engineering for Microstructure–Property Mapping in Organic Photovoltaics

Premium Partner

Marktübersichten