Virtual Laboratory Enabled Constitutive Modelling of Dual Phase Steels
Wencheng Liu,Xinghao Li,Ming Liu,Hao Cui,Jia Huang,Yong Pang,Jun Ma
DOI: https://doi.org/10.1016/j.ijplas.2024.103930
IF: 8.5
2024-01-01
International Journal of Plasticity
Abstract:Accurate and efficient characterization and modelling of deformation responses are critically essential in the development of advanced metal forming processes. This work presents a virtual laboratory enabled constitutive framework for modelling complex deformation responses of dual phase (DP) steels under complex stress states, based on multi-phase full-field crystal plasticity (CP) and advanced phenomenological modelling. In the CP based virtual testing method, kinematic hardening and degradation of elastic modulus are modelled in particular to improve the capability for describing evolving microstructure induced mechanical responses of DP steels. The workflow of this framework is completely built and numerically implemented, including (i) representative volume element (RVE) generation based on microstructure characterisation, (ii) identification of multiphase CP model parameters, (iii) prediction of elastic modulus degradation, (iv) prediction of yield stresses and plastic potentials in uniaxial tension and biaxial tension with various stress ratios, and (v) prediction of stress–strain curves in reverse tension compression. Using DP780 and DP980 as case materials, the corresponding physical experiments are conducted to verify the accuracy of the virtual tests, showing a good agreement between the virtual and experimental approaches. Both experimental and virtual tests are used to calibrate advanced phenomenological constitutive models that include non-linear elasticity, anisotropic yielding, and kinematic hardening. The calibrated models are implemented into the finite element (FE) codes to predict complex deformation and springback behaviours of DP780 and DP980 sheets in U-bending processes. In comparison with forming experiments of U-channel parts, the virtually calibrated models are validated and show good performance in predicting deformation and springback behaviours, providing a high capability for process analysis. The findings support that the virtual laboratory enabled modelling approach could be a substitute for extensive, expensive, and hard-to-access physically mechanical experiments required in the model calibration for a more effective and efficient analysis of metal forming processes.