Study On The Virtual Field Methods For Elastoplasticity And Damage Parameter Inversion Of Laser Welds

In large components manufacturing for aerospace equipment,as a means of connecting aluminum alloy materials,the laser welding additive manufacturing can greatly reduce the structural weight,reduce production cost and improve welding efficiency.At the same time,as an important process for the integrated manufacturing of aviation aluminum alloy siding,the advanced laser welding method has become a key technology in the field of the additive manufacturing in Made in China 2025.However,the thermal gradient effect of laser welding inevitably affects the material microstructure of the weld zone,resulting in degradation of the mechanical load bearing properties of the welded structure.Therefore,the study of the mechanical evolution behavior of aviation aluminum alloy laser welding not only helps to select the welding parameters and to optimize the welding process to reduce the thermal deformation,but also provides the reliable material parameters for the mechanical calculation and the optimization design of the laser welded structure.The randomness and mechanical stability of the laser welding additive manufacturing are the important factors influencing the overall manufacturing quality of the panel.The experimental characterization of the mechanical behavior of laser welds and the constitutive parameter inversion of weld materials are both the key stages in understanding the mechanism of degradation of weld mechanical properties and optimizing the laser welding process.In this thesis,the experimental measurement of the uneven deformation around the aluminum alloy 6061 laser welded joint,the field data mining of the constitutive parameters of the material,the elastoplastic analysis and the failure mechanism of the welds are studied to understand the linear and nonlinear behavior of the laser weld.The main research contents and results are as follows.The invertible second-order deformation operator is constructed by introducing the auxiliary conjugated displacement functions,which can reflect the more complex uneven deformation and the large deformation.And a new second-order inverse compositional matching algorithm(IC-GN2)for the digital image correlation is proposed.The numerical analysis of the simulated speckle images with the large deformation(10%)and the sine uneven deformation respectively confirms that the proposed algorithm has higher precision than the forward combination matching algorithm(FC-GN2)and the previous 1C-GN2.The tensile test of large deformation rubber shows that the new IC-GN2 algorithm has the best performance under the large deformation of about 7.5%,which can provide a full field deformation measurement method for the local deformation behavior of the laser weld and the evolution of the nonlinear behavior of the material.Based on the plastic kinematic constitutive behavior of the aluminum alloy,a predictive corrected stress reconstruction algorithm of kinematic hardening is proposed,which is successfully applied to the nonlinear simplex search in the virtual field method(VFM).The inversion analysis of the prior field data by finite element simulation proves that the method is effective and the noise sensitivity analysis is carried out.The three-dimensional digital image correlation system is used to obtain the uneven deformation field sequence of the aluminum alloy 6061 laser welded specimen during the stretching process.Considering the influence of residual stress on the inversion results,the elastic modulus spectrum and Poisson's ratio spectrum of the aluminum alloy laser welded joint material are identified in the elastic loading stage using the VFM based on the incremental virtual work equation.Then the nonlinear VFM is used to invert the plastic constitutive parameter spectrum,and the stress-strain relationship in each subzone is established.The failure behavior of plastic deformation is explained by analysis of the plastic deformation evolution rate.Based on the idea of return mapping iteration,a univariate return mapping method for isotropically hardening elastoplastic stress increrment updating is proposed,and it has the global convergence and the first-order convergence speed,which can guarantee the stability and search speed of the nonlinear simplex search in the VFM.Through the inversion of the prior field data and the noise sensitivity analysis by the finite element simulation,it is proved that the nonlinear VFM using the stress reconstruction method can achieve higher constitutive parameter identification accuracy.Furthermore,combining the Lemaitre elastoplastic damage coupling model,the inversion interval of the constitutive model parameters of the welding material is extended from the traditional elastoplastic stage to the elastoplastic damage coupling stage by the proposed nonlinear VFM,and the spectrums of elastoplastic parameters and damage parameters of the aluminum alloy 6061 laser welded joint are identified.Finally,the inversion results,combined with the observation of the microscopic crystal structure of the weld,explain the plastic damage evolution behavior of the laser weld.