Evaluation and modeling of effect of strain rate on double yielding behavior of semi-crystalline polymer

Semi-crystalline polymer typically exhibits complex viscoelastic-viscoplastic behavior including the double-yielding (DY) phenomenon. Uniaxial tensile tests at different strain rates were performed using PA specimens obtained at different annealing temperatures to evaluate the effect of the strain rate on the DY phenomenon. The slope of the stress-strain curve after the first yield and the distance between the first and second yields increased with decreasing strain rate. In contrast, the maximum stress observed during the DY phenomenon was independent of the strain rate, and uniform deformation continued for a longer strain range at lower strain rates. The viscoelastic-viscoplastic transient network model was generalized to reproduce the experimentally observed time-dependent DY phenomenon. The proposed method requires over 20 material parameters to reproduce complex mechanical behavior including time-dependent DY phenomenon. To obtain parameters accurately reproduce the experimental results, we introduced machine-learning supported Newton-Raphson identification process. The FEM simulation results with material parameters identified by the proposed method could accurately reproduce the experimentally observed mechanical responses.