A deep learning driven uncertain full‐field homogenization method

This work is directed to uncertainty quantification of homogenized effective properties composite materials with a complex, three dimensional microstructure. The uncertainties arise in the material parameters single constituents as well fiber volume fraction. They are taken into account by multivariate random variables. Uncertainty carried out an efficient surrogate model based on pseudospectral polynomial chaos expansion and artificial neural networks, which trained fast Fourier transformation homogenization method. numerical example deals comparison presented method Monte Carlo-type simulation for uncertain spherical inclusions matrix material.