Idtitle = "A micromechanical Mean-Field Homogenization surrogate for the stochastic multiscale analysis of composite materials failure"journal = International Journal for Numerical Methods in Engineeringyear = 2023volume = 124pages = 5200-5262doi = 10.1002/nme.7344authors = "Calleja, Juan Manuel and Wu Ling, and Nguyen, Van-Dung and Noels, Ludovic" If you use these data or model, we would be grateful if you could cite this above paper SoftwareRequires GMSH and Python 3 with packages numpy, matplotlib, sklearn (scikit-learn), os, pickle, scipy, pandas, cvs, math, seaborn.Each folder contains readme that will help the user to navigate through the data. To run the model you need the open source code https://gitlab.onelab.info/cm3/cm3Libraries but you need to request access to cm3MFH as well Directories Main: Contains fast and easy access to the plots presented in the paper. The readme contained in this plot specifies the plots that are run with each code. 1_SVE_Generator:Contains the files needed for the generation of the SVE, the statistical properties of the microstructure, and PLY samples for the full-field simulations, as well as the used samples 2_Full_Field: contains the extracted data from the FF composite realizations, as well as the used random SVE geometries. 3_Identification: Contains the identification code to find the effective parameters for each SVE realization as well as the obtained identification results. 4_Generator: Contains the generated set of parameters for the 25 and 45 micrometer squared SVEs as well as the codes for the new data generation, the file with the generated data and the plots related with the MF-ROM random parameters and their cross-relations shown in Sections 2.5.2, 3.2.3 and 4. 5_Tests: Contains all the information concerning the tests used for the verification of the MF-ROM and the ply and experimental compression results. MFH_vs_FF: Allows to easily test the inverse identification process through the use of random SVEs and verify the performance of the identified MFH parameters against its full-field counterpart. Plot of figures Figure 9 : Run "python plot_Gc.py" which can be found in folder Main/Full_Field_EnergyFigure 10: Run "python3 PDF_HIST_Gc.py", which can be found in folder Main/HistogramsFigure 23: Run "python3 plot.py" which can be found in folder Main/MFH_FF_ComparisonFigure 24: Run "python3 plot.py" which can be found in folder Main/MFH_FF_ComparisonFigure 27: Run "python3 Correlation_Graphs_25.py contained in folder Main/Distributions_25_Micrometer_SVEFigure 29: Run "python3 PDF_HIST.py" which can be found in folder Main/HistogramsFigure 30: To obtain the data used in this figure, run "python3 DistanceCorrelation_25.py" which can be found in folder /4_GeneratorFigure 31: To obtain the data used in this figure, run "python3 DistanceCorrelation_45.py" which can be found in folder /4_GeneratorFigure 32: Run "python3 Correlation_Graphs_25.py" which can be found in folder Main/Distributions_25_Micrometer_SVEFigure 33: Run "python3 Correlation_Graphs_25.py" which can be found in folder Main/Distributions_25_Micrometer_SVEFigure 34: Run "python3 Correlation_Graphs_25.py" which can be found in folder Main/Distributions_25_Micrometer_SVEFigure 36: Run "python3 plot_New.py" which can be found in folder Main/PlyTestsFigure 46: Run "python3 plot_Test.py" which can be found in folder Main/CompressionExperimentFigure B3: Run "python3 MicroStrAna.py" which can be found in folder Main/MicroStructStatisticsFigure B4: Run "python3 MicroStrAna.py" which can be found in folder Main/MicroStructStatisticsFigure D5: Run "python3 PDF_HIST_B.py" which can be found in folder Main/HistogramsFigure D6: Run "python3 PDF_HIST_B.py" which can be found in folder Main/HistogramsFigure D7: Run "python3 PDF_HIST_B.py" which can be found in folder Main/HistogramsFigure D8: Run "python3 PDF_HIST_B.py" which can be found in folder Main/HistogramsFigure D9: Run "python3 PDF_HIST_B.py" which can be found n folder Main/HistogramsFigure D10: Run "python3 PDF_HIST_B.py" which can be found in folder Main/HistogramsFigure D11: Run "python3 PDF_HIST_B.py" which can be found in folder Main/HistogramsFigure D12: Run "python3 PDF_HIST_B.py" which can be found in folder Main/HistogramsFigure D13: Run "python3 PDF_HIST_B.py" which can be found in folder Main/HistogramsFigure D14: Run "python3 PDF_HIST_B.py" which can be found in folder Main/HistogramsFigure E15: Run "python3 Correlation_Graphs_45.py" which can be found in folder Main/Distributions_45_Micrometer_SVEFigure E16: Run "python3 Correlation_Graphs_45.py" which can be found in folder Main/Distributions_45_Micrometer_SVEFigure E17: Run "python3 Correlation_Graphs_45.py" which can be found in folder Main/Distributions_45_Micrometer_SVEFigure F18: Run "python3 plot_Convergence_25.py" which can be found in folder Main/ConvergenceFigure F19: Run "python3 plot_Convergence_45.py" which can be found in folder Main/Convergence

The authors would like to acknowledge the financial support from F.R.S.-F.N.R.S.as this publication benefits from the support of the Walloon Region within the framework of a FRIA grant. The authors would also like to acknowledge the supercomputing facilities of the Consortium des Équipements de Calcul Intensif en Federation Wallonie Bruxelles (CÉCI) that were funded by the FRS-FNRS that were made available to us.