PSI - Issue 31

David Liović et al. / Procedia Structural Integrity 31 (2021) 86– 91 David Liovi ć et al. / Structural Integrity Procedia 00 (2019) 000–000

88

3

According to (1), it can be concluded that the R - O model predicts a certain, but exceedingly small, plastic strain even at lower values of stress than the yield strength, which in reality is not the case for most metallic materials. In addition, there exists various formulations of the R - O model that may provide slightly different results.

Fig. 1. (a) Monotonic true stress – strain curve of annealed SLM-ed Ti6Al4V alloy; (b) Monotonic R – O parameters for annealed SLM-ed Ti6Al4V alloy,

After determination of initial values of monotonic R – O parameters (Fig. 2. a) and later also cyclic R – O parameters, calibration based on the generalized reduced gradient method is used in order to minimize mean absolute error ( MAE ) to get a better fit between true strain values of R – O curves and experimental stress – strain curves (Fig. 2. b). It should be noted that the monotonic total true strain can be obtained directly by using an analytical approach to equation (1), whereas the relevant iterative methods are required to determine true stress value. The same case applies to determining the cyclic total true strain and stress values.

Fig. 2. (a) Initial R – O monotonic true stress – strain curve (black) and true stress – strain curve (green); (b) Calibrated R – O monotonic true stress – strain curve (black) and true stress – strain curve (green) In the case of modeling cyclic elastoplastic behavior by using R – O material model, true total strain amplitude a ε can be determined according to: 1 ' a a a e,a p,a ' n E K σ σ ε ε ε   = + = +     , (2) where, e,a ε is true elastic strain amplitude, p,a ε is true plastic strain amplitude, ' K is cyclic strength coefficient, ' n is cyclic strain hardening exponent and a σ is true stress amplitude.