Data-driven computational method for growth-induced deformation problems of soft materials.

This paper presents a data-driven solver (DDS) for the growth-induced deformation problems of soft materials. Contrary to the over-reliance of traditional numerical methods on material constitutive models, the DDS requires only a discrete material dataset of stress–strain pairs to describe the stress–strain relationship. With the multiplicative decomposition of the gradient tensor, the growth effects are introduced. The growth-induced deformation problems are then solved as the pre-strain problems. To overcome the difficulty in determining the growth-induced pre-stress without an explicit constitutive model, a strain-driven searching scheme is thus proposed. Several numerical examples demonstrate the robustness and its good performance of the proposed DDS. In particular, the reason for choosing the strain-driven searching scheme is also illustrated according to the numerical example results. [ABSTRACT FROM AUTHOR]