Adaptive finite element convergence analysis of AT1 phase-field model for quasi-static fracture in strain-limiting solids

Abstract

This research rigorously investigates the convergence of adaptive finite element methods for regularized variational models of quasi-static brittle fracture in elastic solids. We specifically examine a novel Ambrosio-Tortorelli (AT1) phase-field model within the framework of elasticity theories, particularly for material models characterized by an algebraically nonlinear stress-strain relationship. Two distinct and novel adaptive mesh refinement algorithms, underpinned by robust local error indicators, were introduced to efficiently solve the underlying nonlinear energy minimization problem. A detailed convergence analysis was conducted on the sequences of minimizers produced by these strategies. Our findings rigorously demonstrate that the minimizer sequences from the first adaptive algorithm achieve convergence to a predefined tolerance. Crucially, the second algorithm is proven to generate inherently convergent sequences, thereby eliminating the need for an explicit stopping criterion. The practical effectiveness of this proposed adaptive framework is thoroughly validated through extensive numerical simulations. A case study involving an edge crack in an elastic body, governed by an algebraically nonlinear strain-limiting relationship and subjected to anti-plane shear-type loading, is presented. Critical comparisons of the energy components-bulk, surface, and total-showcase the superior performance of both adaptive algorithms.