From strain to stress using full-field data: Computationally efficient stress reconstruction
Abstract. Conventional stress reconstruction based on full-field strain measurements presents a major computational burden, especially when using standard implicit stress integration methods. This presents a notable challenge for inverse identification methods used to characterize the plasticity of metallic materials, particularly those reliant on stress reconstruction, such as the nonlinear sensitivity-based Virtual Fields Method (VFM). To reduce the computational effort, the full-field strain data are usually spatially and temporally down-sampled. However, for metals subject to nonlinear strain paths, this practice can lead to errors in the resulting stress states and compromise the accuracy of the nonlinear VFM. In this work, we introduce a highly efficient explicit stress reconstruction algorithm to reduce the computational challenges of repeated stress reconstruction which can be utilized in inverse identification methods such as nonlinear VFM.