PhySRNet: Physics informed super-resolution network for application in computational solid mechanics

Abstract

Traditional numerical approaches have been successfully used to model mechanical behavior of heterogeneous materials (composites, multicomponent alloys, and polycrystals) widely used in industrial applications. However, these methods require a fine mesh resulting in computationally expensive and time-consuming calculations. The physics-informed deep-learning based super-resolution framework (PhySRNet) introduced in this paper is aimed at overcoming this computational challenge. PhySRNet enables reconstruction of high-resolution solution fields from their low-resolution counterparts without requiring labeled data, thereby allowing researchers to run their numerical simulations on a coarse mesh. Through an illustrative example, we demonstrate that the super-resolved fields match the accuracy of an advanced numerical solver running at 400 times the coarse mesh resolution and satisfy the (highly nonlinear) governing laws. The approach opens the door to applying machine learning and traditional numerical approaches in tandem to reduce computational complexity and accelerate scientific discovery and engineering design.