Constitutive modeling of non-ideal isotropic materials based on a novel theoretical framework

Abstract

If a plastic constitutive relation is used to simulate a metal forming process, the main concern should be its ability to predict the plastic strain parameters. At present, constitutive modelling basically relies on the properties of stress parameters, and emphasizes on how to improve the prediction accuracy of initial yield and subsequent hardening stresses. Although, corresponding to the anisotropic materials, the experimental data of plastic strain ratios are also used, the experimental plastic strains data obtained from all related experiments, in addition to the one as the equivalent plastic strain, are not employed at all. The lack of the most important experimental data in the constitutive model will inevitably weaken its ability to predict the plastic strain parameters which will determine the final shape of formed parts.To make a constructed constitutive relation that can fully reflect the experimental characteristics of materials, a new theoretical framework of constitutive model is proposed, and the application of experimental plastic strain data can be achieved with the introduction of a plastic strain-increment function. To illustrate the importance of the application of plastic strain-increment function to improve the predictive accuracy of plastic strain-increments, a new constitutive model is constructed and combined with a large amount of experimental data to verify it. Experimental verifications were carried out by some sheet metals with different anisotropic properties under uniaxial and equi-biaxial tension stress states, and also deal with multiple sets of experimental data obtained under complex tension-tension and tension-compression stress states. Discussions also include the results predicted by the models constructed based on the classical theory. Calculation examples show, the stronger the anisotropic properties of material, the greater the prediction deviation of plastic strains generated by the classical-theory-based models. Moreover, such prediction deviations will gradually increase with the accumulation of predicted plastic strain-increments. Predictive ability of the new constitutive relation is completely unimpacted by the increased degree of anisotropic features of materials. Predicted results are in good agreement with all experimental data.