Adaptive implicit–explicit method for robust and efficient failure analysis of composite materials

Abstract

Despite numerous methods devised for the analysis of composite materials, substantial challenges persist in modeling the progressive failure process. Specifically, the damage-induced strain softening may cause convergence difficulty in implicit procedures, whereas lack of accuracy and restriction on time-step length are problems of concern for explicit formulations. Herein, we develop an adaptive implicit–explicit method (A-IEM) to enhance the robustness and efficiency of computational failure models. Initially, implicit algorithms are applied for non-linear analysis, during which critical nodes that hinder the convergence rate are identified through a concurrent monitoring algorithm. Upon fatal non-convergence, explicit constitutive law is locally enforced on the corresponding material points to evaluate the damage progression within current increment, and reversion to implicit scheme is subsequently enabled by A-IEM after resolving the numerical issues. The performance of A-IEM against conventional methods is verified via standard numerical examples and further demonstrated in the failure modeling of composite laminates.