| In the background of lightweight,high strength aluminum alloy sheets have been widely used in aviation,aerospace,automobiles,ships and other fields due to their small density,high specific strength and strong toughness.However,there are problems of poor plasticity and obvious springback in the forming process of high strength aluminum alloy sheet parts,especially for part with large flat,small curvature and complex surface.Both springback prediction method and springback prediction accuracy are the main problems in practical engineering.Finite element simulation(FEA)is the main method for stress analysis and springback prediction of sheet forming,and the key to its accuracy lies in whether the stress distribution of sheet bending forming and the nonlinear elasticity problem in sheet springback can be accurately described.In response to this problem,this paper selects AA7075 sheet,combined with the forming process,takes its T6 state and solid solution state as the research object.The anisotropic constitutive model is constructed under the guidance of FEA theory and elastoplastic constitutive theory.The influence of the nonlinear elastic behavior and its experimental solution strategy on the prediction accuracy of springback is emphatically studied,in order to improve the accuracy of springback prediction.Uniaxial tensile and shear material performance test,combined with the DIC technology,are designed to calibrate the elastoplastic mechanical parameters and the anisotropy coefficient change law of AA7075 sheet.Based on this,the evolution of the parameters of the Yld2000-2d yield criterion is described,and the Yld2000-2d distortion enhancement model and incremental expression are established.Based on the Abaqus FE platform,the secondary development of the constitutive subroutine is carried out.The fully implicit backward Euler algorithm is used to derive the stress update method and the consistent tangent modulus expression,then the FE embedding of the model is realized.Three mechanically-measured tests,cyclic loading test,preloading-bending test and bending-unloading test,are separately implemented to study the degradation of elastic modulus.Results shows that the elastic modulus of AA7075 sheet decreased rapidly with the increase of plastic deformation and then gradually stabilized.The elastic modules decreased about 20.4%,8.3% and 31.1% under the above three test methods respectively of AA7075-T6 sheet;and 16.6%,11.1%,23.5% of AA7075-gurong sheet.The solution treatment reduces the degrading effect of the elastic modulus obviously.The strain-driven elastic modulus degradation law is embedded in the constitutive subroutine through the secondary development of Abaqus,and the springback prediction of three-point bending is carried out.The analysis shows that the constant elastic modulus underestimated the rebound amount of the specimen.The variable elastic modulus model can significantly improve the prediction accuracy of the rebound under large deformation.The increase in accuracy comes from the unloading of more elastic strains.Compared with the traditional test method,the bending-unloading test method designed in this paper significantly improves the springback prediction accuracy.The bending-unloading elastic modulus can accurately predict the three-point bending springback of the AA7075 sheet after solution treatment.The prediction errors of this model for the opening springback angles of T6 and solid solution samples are 3% and 1%,respectively,and can accurately predict the three-point bending springback of the AA7075 sheet after solution treatment.The stamping test of the C-beam specimen was designed to combine with the FEM to analyze the torsional springback characteristics and springback prediction.Through the forming analysis shows that the springback of the sample opening and the twist springback are dominated by the transverse and longitudinal stress fields,respectively.The stress state of the arch core formed by bending is different from that of the arch back,which causes the springback to generate opposite unbalanced moments and changes in different curvature radii.That are the main reasons for the torsional springback.Through the springback prediction shows that the prediction accuracy of the four variable elastic modulus models for the opening springback and torsional springback of the C-beam is consistent with the three-point bending results.The solution treatment suppresses the springback and significantly reduces the difficulty of springback prediction.Combined with the bend-unload model,the springback and can be basically accurately predicted. |
