Improvement of the prediction accuracy of Hill48 yield criterion with novel parameter identification method

İngilizce (OpenAlex)

Purpose The aim of the present study is to improve the prediction capability of the Hill48 orthotropic yield criterion by applying the novel calibration method. Design/methodology/approach Two different coefficient identification methods based on numerical optimization techniques were proposed in the study. In the first method, total effect of the yield stresses and Lankford coefficients in main directions were considered, whereas in the second method, the effect of each yield stresses and Lankford coefficient were independently taken into account. The mentioned two methods were applied to describe the anisotropic behavior of a commercial pure titanium sheet in order to investigate their effects on the prediction accuracy of Hill48 criterion and the cup drawing test was selected as a benchmark study. Findings From the comparisons it was determined that only the latter method could successfully predict both earing profile and cup height of the deep drawn cup. In this study, successful results were obtained by improving the prediction performance of Hill48 using the new proposed optimization-based calibration methods. Originality/value In this study, a novel parameter calibration method has been developed and applied to model anisotropic behavior of commercial pure titanium sheet, which exhibits strongly anisotropic behavior. After comparison of the numerical and experimental results it is seen that the developed method has accurately modelled the planar anisotropy and a significant enhancement is provided when compared to the classical analytical methods.