Machine learning approaches for elastic localization linkages in high-contrast composite materials

There has been a growing recognition of the opportunities afforded by advanced data science and informatics approaches in addressing the computational demands of modeling and simulation of multiscale materials science phenomena. More specifically, the mining of microstructure–property relationships by various methods in machine learning and data mining opens exciting new opportunities that can potentially result in a fast and efficient material design. This work explores and presents multiple viable approaches for computationally efficient predictions of the microscale elastic strain fields in a three-dimensional (3-D) voxel-based microstructure volume element (MVE). Advanced concepts in machine learning and data mining, including feature extraction, feature ranking and selection, and regression modeling, are explored as data experiments. Improvements are demonstrated in a gradually escalated fashion achieved by (1) feature descriptors introduced to represent voxel neighborhood characteristics, (2) a reduced set of descriptors with top importance, and (3) an ensemble-based regression technique.

[1]  Bin Wen,et al.  Computing mechanical response variability of polycrystalline microstructures through dimensionality reduction techniques , 2010 .

[2]  Yuksel C. Yabansu,et al.  Understanding and visualizing microstructure and microstructure variance as a stochastic process , 2011 .

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

[4]  John Hinde,et al.  Statistical Modelling in R , 2009 .

[5]  H. Egawa,et al.  Measurement of three dimensional stress and modeling of stress induced migration failure in aluminium interconnects , 1990, 28th Annual Proceedings on Reliability Physics Symposium.

[6]  J. R. Quinlan.  Learning With Continuous Classes , 1992 .

[7]  H. Garmestani,et al.  Microstructure design of a two phase composite using two-point correlation functions , 2004 .

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

[9]  Kim F. Ferris,et al.  Materials informatics : Fast track to new materials , 2007 .

[10]  Surya R. Kalidindi,et al.  Formulation and calibration of higher-order elastic localization relationships using the MKS approach , 2011 .

[11]  A. Molinari,et al.  A STATISTICAL FORMULATION OF VISCOPLASTIC BEHAVIOR IN HETEROGENEOUS POLYCRYSTALS , 1989 .

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

[13]  Johannes Hötzer,et al.  Analytics for microstructure datasets produced by phase-field simulations , 2016 .

[14]  S. Torquato,et al.  Random Heterogeneous Materials: Microstructure and Macroscopic Properties , 2005 .

[15]  Phillip B. Messersmith,et al.  Bioinspired antifouling polymers , 2005 .

[16]  G. Proust,et al.  Procedures for construction of anisotropic elastic–plastic property closures for face-centered cubic polycrystals using first-order bounding relations , 2006 .

[17]  Bernhard Schölkopf,et al.  A tutorial on support vector regression , 2004, Stat. Comput..

[18]  Surya R. Kalidindi,et al.  Microstructure Informatics Using Higher-Order Statistics and Efficient Data-Mining Protocols , 2011 .

[19]  Surya R. Kalidindi,et al.  Microstructure informatics using higher-order statistics and efficient data-mining protocols , 2011 .

[20]  Jitesh H. Panchal,et al.  Key computational modeling issues in Integrated Computational Materials Engineering , 2013, Comput. Aided Des..

[21]  Krishna Rajan,et al.  Principal component analysis and dimensional analysis as materials informatics tools to reduce dimensionality in materials science and engineering , 2009, Stat. Anal. Data Min..

[22]  Surya R. Kalidindi,et al.  Application of data science tools to quantify and distinguish between structures and models in molecular dynamics datasets , 2015, Nanotechnology.

[23]  Pascal Doumalin,et al.  Comparison between X-ray micro-computed tomography and optical scanning tomography for full 3D strain measurement by digital volume correlation , 2008 .

[24]  S. Kalidindi,et al.  Novel microstructure quantification framework for databasing, visualization, and analysis of microstructure data , 2013, Integrating Materials and Manufacturing Innovation.

[25]  D. Fullwood,et al.  Optimized structure based representative volume element sets reflecting the ensemble-averaged 2-point statistics , 2010 .

[26]  B. Adams,et al.  Microstructures by design: linear problems in elastic–plastic design , 2004 .

[27]  David L. McDowell,et al.  Linking phase field and finite element modeling for process-structure-property relations of a Ni-base superalloy , 2012 .

[28]  Alfredo Valarezo,et al.  Partnership for accelerated insertion of new technology: case study for thermal spray technology , 2013, Integrating Materials and Manufacturing Innovation.

[29]  Thomas Böhlke,et al.  Representation of Hashin–Shtrikman bounds of cubic crystal aggregates in terms of texture coefficients with application in materials design , 2014 .

[30]  John Tyson,et al.  Pull-field dynamic displacement and strain measurement using advanced 3D image correlation photogrammetry: Part 1 , 2003 .

[31]  Surya R. Kalidindi,et al.  Multi-scale modeling of the elastic response of a structural component made from a composite material using the materials knowledge system , 2012 .

[32]  Surya R. Kalidindi,et al.  Representation and calibration of elastic localization kernels for a broad class of cubic polycrystals , 2015 .

[33]  David T. Fullwood,et al.  Microstructure Sensitive Design for Performance Optimization , 2012 .

[34]  Surya R. Kalidindi,et al.  Thermo-Elastic Localization Relationships for Multi-Phase Composites , 2010 .

[35]  Surya R. Kalidindi,et al.  A Novel Framework for Building Materials Knowledge Systems , 2010 .

[36]  David T. Fullwood,et al.  Elastic properties closures using second-order homogenization theories: Case studies in composites of two isotropic constituents , 2006 .

[37]  Surya R. Kalidindi,et al.  Fast computation of first-order elastic–plastic closures for polycrystalline cubic-orthorhombic microstructures , 2007 .

[38]  Ian H. Witten,et al.  Induction of model trees for predicting continuous classes , 1996 .

[39]  S. Torquato.  Random Heterogeneous Materials , 2002 .

[40]  Dan Backman,et al.  Integrated computational materials engineering: A new paradigm for the global materials profession , 2006 .

[41]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[42]  Surya R. Kalidindi,et al.  Data science and cyberinfrastructure: critical enablers for accelerated development of hierarchical materials , 2015 .

[43]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[44]  Surya R. Kalidindi,et al.  A data-driven approach to establishing microstructure–property relationships in porous transport layers of polymer electrolyte fuel cells , 2014 .

[45]  A. Belvin,et al.  Application of two-point probability distribution functions to predict properties of heterogeneous two-phase materials , 2009 .

[46]  Rajen Dinesh Shah,et al.  Statistical modelling , 2015 .

[47]  Andrea Widener,et al.  Materials Genome Initiative , 2014 .

[48]  David T. Fullwood,et al.  A new spectral framework for establishing localization relationships for elastic behavior of composites and their calibration to finite-element models , 2008 .

[49]  David L. McDowell,et al.  Concurrent design of hierarchical materials and structures , 2008 .

[50]  S. Ahzi,et al.  Statistical continuum theory for large plastic deformation of polycrystalline materials , 2001 .

[51]  Surya R. Kalidindi,et al.  A new framework for computationally efficient structure–structure evolution linkages to facilitate high-fidelity scale bridging in multi-scale materials models , 2011 .

[52]  Krishna Rajan,et al.  Invited review: Data mining and informatics for crystal chemistry: establishing measurement techniques for mapping structure–property relationships , 2009 .

[53]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[54]  Surya R. Kalidindi,et al.  Multi-scale modeling of elastic response of three-dimensional voxel-based microstructure datasets using novel DFT-based knowledge systems , 2010 .