Big data for microstructure‐property relationships: A case study of predicting effective conductivities

The analysis of big data is changing industries, businesses and research as large amounts of data are available nowadays. In the area of microstructures, acquisition of (3-D tomographic image) data is difficult and time-consuming. It is shown that large amounts of data representing the geometry of virtual, but realistic 3-D microstructures can be generated using stochastic microstructure modeling. Combining the model output with physical simulations and data mining techniques, microstructure-property relationships can be quantitatively characterized. Exemplarily, we aim to predict effective conductivities given the microstructure characteristics volume fraction, mean geodesic tortuosity, and constrictivity. Therefore, we analyze 8119 microstructures generated by two different stochastic 3-D microstructure models. This is—to the best of our knowledge—by far the largest set of microstructures that has ever been analyzed. Fitting artificial neural networks, random forests and classical equations, the prediction of effective conductivities based on geometric microstructure characteristics is possible. © 2017 American Institute of Chemical Engineers AIChE J, 63: 4224–4232, 2017

[1]  V. Schmidt,et al.  Quantitative relationships between microstructure and effective transport properties based on virtual materials testing , 2014 .

[2]  V. Schmidt,et al.  STOCHASTIC 3D MODELING OF AMORPHOUS MICROSTRUCTURES - A POWERFUL TOOL FOR VIRTUAL MATERIALS TESTING , 2016 .

[3]  Mohammad Bagher Menhaj,et al.  Training feedforward networks with the Marquardt algorithm , 1994, IEEE Trans. Neural Networks.

[4]  Ashley A. White.  Big data are shaping the future of materials science , 2013 .

[5]  P. Gasser,et al.  Fundamental relationships between 3D pore topology, electrolyte conduction and flow properties : towards knowledge-based design of ceramic diaphragms for sensor applications , 2016 .

[6]  Eric R. Ziegel,et al.  The Elements of Statistical Learning , 2003, Technometrics.

[7]  M. Doyle,et al.  Modeling of Galvanostatic Charge and Discharge of the Lithium/Polymer/Insertion Cell , 1993 .

[8]  Bernard Widrow,et al.  Improving the learning speed of 2-layer neural networks by choosing initial values of the adaptive weights , 1990, 1990 IJCNN International Joint Conference on Neural Networks.

[9]  Volker Schmidt,et al.  Predicting Effective Conductivities Based on Geometric Microstructure Characteristics , 2016 .

[10]  P. Gasser,et al.  3D Microstructure Effects in Ni-YSZ Anodes: Prediction of Effective Transport Properties and Optimization of Redox Stability , 2015, Materials.

[11]  Nobuhide Kasagi,et al.  Numerical Assessment of SOFC Anode Polarization Based on Three-Dimensional Model Microstructure Reconstructed from FIB-SEM Images , 2010 .

[12]  Antoinette Tordesillas,et al.  Machine learning framework for analysis of transport through complex networks in porous, granular media: A focus on permeability. , 2016, Physical review. E.

[13]  W. Bessler,et al.  Low-temperature charging of lithium-ion cells part I: Electrochemical modeling and experimental investigation of degradation behavior , 2014 .

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

[15]  Pierre Soille,et al.  Morphological Image Analysis: Principles and Applications , 2003 .

[16]  K. Kurzydłowski,et al.  The influence of pore size variation on the pressure drop in open-cell foams , 2015 .

[17]  I. Molchanov.  Statistics of the Boolean Model for Practitioners and Mathematicians , 1997 .

[18]  Surya R. Kalidindi,et al.  Materials Data Science: Current Status and Future Outlook , 2015 .

[19]  Andy Liaw,et al.  Classification and Regression by randomForest , 2007 .

[20]  Clemens Bechinger,et al.  Direct relations between morphology and transport in Boolean models. , 2015, Physical review. E, Statistical, nonlinear, and soft matter physics.

[21]  Lorenz Holzer,et al.  Contradicting Geometrical Concepts in Pore Size Analysis Attained with Electron Microscopy and Mercury Intrusion , 2008 .

[22]  F. Marone,et al.  Determination of Material Properties of Gas Diffusion Layers: Experiments and Simulations Using Phase Contrast Tomographic Microscopy , 2009 .

[23]  Volker Schmidt,et al.  A New Approach to Model‐Based Simulation of Disordered Polymer Blend Solar Cells , 2012 .

[24]  Koster,et al.  Submitted to the Annals of Applied Statistics SPATIAL MODELING OF THE 3 D MORPHOLOGY OF HYBRID POLYMER-ZNO SOLAR CELLS , BASED ON ELECTRON TOMOGRAPHY DATA , 2011 .

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

[26]  Eugene E. Petersen,et al.  Diffusion in a pore of varying cross section , 1958 .

[27]  B. Münch,et al.  The influence of constrictivity on the effective transport properties of porous layers in electrolysis and fuel cells , 2013, Journal of Materials Science.

[28]  Peter Dalgaard,et al.  R Development Core Team (2010): R: A language and environment for statistical computing , 2010 .

[29]  K. Kurzydłowski,et al.  Thermal conductivity of metal-graphene composites , 2016 .

[30]  Lorenz Holzer,et al.  Review of FIB tomography , 2012 .