Quantification of similarity and physical awareness of microstructures generated via generative models

Large repositories of microstructure realizations lie at the centre of developing effective structure-property correlations in materials. It is, however important that the microstructure generation procedures not only reconstruct a large number of microstructures in a computationally inexpensive manner but also hold awareness regarding the physical significance of the underlying microstructure. While machine learning techniques are used for computationally efficient microstructure generation, the similarity and physical awareness of the generated microstructures with the ground truth are rarely quantified. In this work, we use a variant of generative adversarial network (GAN) i.e. StyleGAN2, to generate microstructures with varied morphologies from a small dataset of Dual Phase (DP) steels. The similarity between the generated and the original microstructures is quantified using various metrics such as structure similarity index (SSIM), peak signal to noise ratio (PSNR) and signal to noise ratio (SNR). The physical awareness is quantified by comparing the predictions of macroscopic mechanical properties of the GAN generated and original microstructures using a reduced order model. We also qualitatively estimate the learning of the GAN latent space in terms of microstructure morphology. It is observed that there exists a relationship between the microstructure morphological information and the similarity assessment metrics.

Automated summary

Large repositories of microstructure realizations play a crucial role in establishing effective structure-property correlations in materials. However, it is vital to generate microstructures in a computationally efficient manner while maintaining awareness of their physical significance. In this study, we utilize a variant of generative adversarial network (GAN), namely StyleGAN2, to generate microstructures with diverse morphologies from a small dataset. We quantify the similarity between the generated and original microstructures using various metrics and assess their physical awareness by comparing the predictions of macroscopic mechanical properties. Additionally, we qualitatively evaluate the learning of the GAN latent space in terms of microstructure morphology.