An efficient machine learning approach to establish structure-property linkages

Abstract Full-field simulations with synthetic microstructure offer unique opportunities in predicting and understanding the linkage between microstructural variables and properties of a material prior to or in conjunction with experimental efforts. Nevertheless, the computational cost restrains the application of full-field simulations in optimizing materials microstructures or in establishing comprehensive structure-property linkages. To address this issue, we propose the use of machine learning technique, namely Gaussian process regression, with a small number of full-field simulation results to construct structure-property linkages that are accurate over a wide range of microstructures. Furthermore, we demonstrate that with the implementation of expected improvement algorithm, microstructures that exhibit most desirable properties can be identified using even smaller number of full-field simulations.

[1]  V. Kouznetsova,et al.  Deformation behaviour of lath martensite in multi-phase steels , 2016 .

[2]  Victor Picheny,et al.  Noisy kriging-based optimization methods: A unified implementation within the DiceOptim package , 2014, Comput. Stat. Data Anal..

[3]  Cheng Li,et al.  Regret for Expected Improvement over the Best-Observed Value and Stopping Condition , 2017, ACML.

[4]  Yves Rémond,et al.  Effective conductivity in isotropic heterogeneous media using a strong-contrast statistical continuum theory , 2009 .

[5]  Noah H. Paulson,et al.  Reduced-order structure-property linkages for polycrystalline microstructures based on 2-point statistics , 2017 .

[6]  Hyung Keun Park,et al.  Three-dimensional microstructure modeling of particulate composites using statistical synthetic structure and its thermo-mechanical finite element analysis , 2017 .

[7]  William Fuller Brown,et al.  Solid Mixture Permittivities , 1955 .

[8]  Verónica Bolón-Canedo,et al.  A review of microarray datasets and applied feature selection methods , 2014, Inf. Sci..

[9]  David T. Fullwood,et al.  A strong contrast homogenization formulation for multi-phase anisotropic materials , 2008 .

[10]  E. Kröner.  Bounds for effective elastic moduli of disordered materials , 1977 .

[11]  Salvatore Torquato,et al.  Effective stiffness tensor of composite media—I. Exact series expansions , 1997 .

[12]  R. Kondor,et al.  Gaussian approximation potentials: the accuracy of quantum mechanics, without the electrons. , 2009, Physical review letters.

[13]  Somnath Ghosh,et al.  A non-Schmid crystal plasticity finite element approach to multi-scale modeling of nickel-based superalloys , 2016 .

[14]  Surya R. Kalidindi,et al.  Quantification and classification of microstructures in ternary eutectic alloys using 2-point spatial correlations and principal component analyses , 2016 .

[15]  Martin Diehl,et al.  A virtual laboratory using high resolution crystal plasticity simulations to determine the initial yield surface for sheet metal forming operations , 2016 .

[16]  Hyoung-Seop Kim,et al.  Continuum understanding of twin formation near grain boundaries of FCC metals with low stacking fault energy , 2017, npj Computational Materials.

[17]  D. Mahapatra,et al.  A review of micromechanics based models for effective elastic properties of reinforced polymer matrix composites , 2018, Composite Structures.

[18]  F. Roters,et al.  A crystal plasticity model for twinning- and transformation-induced plasticity , 2016 .

[19]  L. Lebensztajn,et al.  Kriging: a useful tool for electromagnetic device optimization , 2004, IEEE Transactions on Magnetics.

[20]  Liguo Zhao,et al.  Realistic microstructure-based modelling of cyclic deformation and crack growth using crystal plasticity , 2016 .

[21]  Nicholas Lubbers,et al.  Inferring low-dimensional microstructure representations using convolutional neural networks , 2016, Physical review. E.

[22]  Xuelong Li,et al.  Structure preserving unsupervised feature selection , 2018, Neurocomputing.

[23]  Philip Lewis,et al.  Geostatistical classification for remote sensing: an introduction , 2000 .

[24]  Alexandros A. Taflanidis,et al.  Sequential approximate optimization for design under uncertainty problems utilizing Kriging metamodeling in augmented input space , 2017 .

[25]  M. Groeber,et al.  DREAM.3D: A Digital Representation Environment for the Analysis of Microstructure in 3D , 2014, Integrating Materials and Manufacturing Innovation.

[26]  Donald R. Jones,et al.  Efficient Global Optimization of Expensive Black-Box Functions , 1998, J. Glob. Optim..

[27]  A. Çeçen,et al.  Development of high throughput assays for establishing process-structure-property linkages in multiphase polycrystalline metals: Application to dual-phase steels , 2017 .

[28]  D. Fullwood,et al.  Microstructure reconstructions from 2-point statistics using phase-recovery algorithms , 2008 .

[29]  Ming Yang,et al.  New algorithms for virtual reconstruction of heterogeneous microstructures , 2018, Computer Methods in Applied Mechanics and Engineering.

[30]  Bülent Yener,et al.  Image driven machine learning methods for microstructure recognition , 2016 .

[31]  Ferat Sahin,et al.  A survey on feature selection methods , 2014, Comput. Electr. Eng..

[32]  A. Amaro,et al.  Modeling and simulation techniques for polymer nanoparticle composites – A review , 2016 .

[33]  Dorin Drignei,et al.  An estimation algorithm for fast kriging surrogates of computer models with unstructured multiple outputs , 2017 .

[34]  Zhen Hu,et al.  Calibration experimental design considering field response and model uncertainty , 2017 .

[35]  Surya R. Kalidindi,et al.  Calibrated localization relationships for elastic response of polycrystalline aggregates , 2014 .

[36]  Yuksel C. Yabansu,et al.  Application of spherical indentation and the materials knowledge system framework to establishing microstructure-yield strength linkages from carbon steel scoops excised from high-temperature exposed components , 2018 .

[37]  H. Garmestani,et al.  Modeling the mechanical behavior of heterogeneous ultrafine grained polycrystalline and nanocrystalline FCC metals , 2018, Mechanics of Materials.

[38]  Nirupam Chakraborti,et al.  A data-driven surrogate-assisted evolutionary algorithm applied to a many-objective blast furnace optimization problem , 2017 .

[39]  K. Müller,et al.  Fast and accurate modeling of molecular atomization energies with machine learning. , 2011, Physical review letters.

[40]  A. Gokhale,et al.  Constraints on microstructural two-point correlation functions , 2005 .

[41]  Yuksel C. Yabansu,et al.  Extraction of reduced-order process-structure linkages from phase-field simulations , 2017 .

[42]  Surya R. Kalidindi,et al.  Data-driven reduced order models for effective yield strength and partitioning of strain in multiphase materials , 2017, J. Comput. Phys..