Cyclic Sheet Metal Test Comparison and Parameter Calibration for Springback Prediction of Dual-Phase Steel Sheets

Springback is an important issue for the application of advanced high-strength steels (AHSS) in the automobile industry. Various studies have shown that it is an effective way to predict springback by using path-dependent material models. The accuracy of these material models greatly depends on the experimental test methods as well as material parameters calibrated from these tests. The present cyclic sheet metal test methods, like uniaxial tension–compression test (TCT) and cyclic shear test (CST), are nonstandard and various. The material parameters calibrated from these tests vary greatly from one to another, which makes the usage of material parameters for the accurate prediction of springback more sophisticated even when the advanced material model is available in commercial software. The focus of this work is to compare the springback prediction accuracy by using the material parameters calibrated from tension–compression test or cyclic shear test, and to further clarify the usage of those material parameters in application. These two types of nonstandard cyclic tests are successfully carried out on a same test platform with different specimen geometries. One-element models with corresponding tension–compression or cyclic shear boundary conditions are built, respectively, to calibrate the parameters of the modified Yoshida–Uemori (YU) model for these two different tests. U-bending process is performed for springback prediction comparison. The results show, for dual phase steel (DP780), the work hardening stagnation is not evident by tension–compression tests at all the prestrain levels or by cyclic shear test at small prestrain γ = 0.20 but is significantly apparent by cyclic shear tests at large prestrain γ = 0.38, 0.52, 0.68, which seems to be a prestrain-dependent phenomenon. The material parameters calibrated from different types of cyclic sheet metal tests can vary greatly, but it gives slight differences of springback prediction for U-bending by utilizing either tension–compression test or cyclic shear test.