A phenomenological constitutive model for deep rolling process simulation of AISI 4140 under high dynamic conditions

Abstract

AISI 4140 alloy steel exhibits complex mechanical behaviors such as the Bauschinger effect, strain rate sensitivity, and strain softening, which traditional constitutive models, like Johnson-Cook model and Chaboche kinematic hardening model, are hard to simultaneously capture. To address this limitation, the present work proposes a phenomenological constitutive model based on the commonly used Johnson-Cook formulation, enhanced by incorporating strain softening, strain rate sensitivity, and the Chaboche kinematic hardening. The developed model is implemented through a vectorised user material (VUMAT) subroutine in Abaqus and validated using numerical simulations of residual stress depth profile induced by deep rolling. It is then applied to simulate deep rolling processes under various forces and rolling speeds. The results demonstrate that the proposed model can accurately reproduce residual stress profiles and internal material loading histories. Among the considered mechanisms, strain rate sensitivity and kinematic hardening are found to be essential for capturing the residual stress evolution, while strain softening plays a secondary role under the investigated conditions.