Multi-physics Fields Simulation And Weld Joint Microstructure/Properties Of 2219-T6 Aluminum Alloy In Electron Beam Welding

Electron beam welding(EBW)has high energy density,high welding speed and narrow heat-affected zone characteristics.This technique has been widely used in aviation,aerospace and other industries.With the development of aerospace and military industries,2219-T6 aluminum alloys are adopted to produce lightweight,high strength structures.Meanwhile,more requirements for the efficiency,control and prediction of the welding process are raised.The process-structure-property relationship has always been the research focus in welding and additive manufacturing.Compared to the experimental methods and data-driven modeling,multi-physics simulation has the advantages of reproducing the dynamic welding process,understanding the physical essence and predicting weld quality.However,most models can not connect effectively the molten pool simulation,microstructure simulation and mechanical properties simulation.For example,the interaction between the electron beam and keyhole surface,the effect of complex fusion line shape on the grain nucleation,the effect of the inhomogeneous microstructure on the mechanical properties are not considered in most studies.In this study,a multiphysics fields simulation framework was established to analyze the relationship between process,microstructure and property.This study was expected to provide a deeper understanding of the physical essence of EBW process and to improve confidence in optimizing and controlling the actual welding process.Firstly,a three-dimensional molten pool model was developed to study the molten pool dynamics during EBW.The model involves an innovative heat source model that considers the broadening effect of metal vapor on the electron beam radius and direct coupling between the electron beam and keyhole surface.Based on this model,the effect of process parameters on the molten pool behavior was studied.Moreover,the mechanism of electron beam scanning on molten pool was revealed.The simulated results were validated by comparing the weld areas and weld dimensions from experiments and simulations.Additionally,the number of porosity defects in the experimental weld seam was examined to prove the rationality of the theoretical analysis.Based on the molten pool model and the theoretical analysis,a Gaussian process regression method for rapidly predicting the weld dimensions and a new method for quantifying the molten pool stability were proposed.Eventually,we developed an effective and rapid method for selecting welding process parameters.Secondly,the cellular automata method was adopted to model the dendrite growth.The effects of temperature gradient and cooling rate on single dendrite growth and multi-physics dendrites growth were discussed.The relationships between the solidification parameters and grain size and segregation are quantitatively described.Taking the temperature field and VOF(Volume of fraction)values in the molten pool as the input data,the microstructure model was developed.The model involves an efficient method to track the second phase and the superposition algorithm of the molten meshes to identify the complex fusion line shapes.The mechanism of microstructure evolution under the effect of molten pool oscillation was elaborated.The effects of the process parameters on the grain morphology,grain size,segregation and second phase were predicted.These predicted results were in good agreement with experimental results.Finally,taking the simulated grain size,second phase and concentration field results as the input data,the yield strength and Vickers hardness distribution of weld were calculated by the analytical models.Various strengthening mechanisms,such as fine grain strengthening,second phase strengthening and solid solution strengthening,were considered.Taking the predicted yield strength as the material parameter and Johnson cook model as the constitutive equation,a finite element model was established to simulate the tensile process and predict the tensile strength.The predicted results were in good agreement with experimental results.In this study,a physics-driven modeling framework was established to connect the molten pool simulation,microstructure simulation and mechanical properties simulation.The improved models tackle many limitations in the existing works and greatly complement the modeling framework.This framework captures the multiphysical coupling phenomena in the melting and solidification process,such as electron beam-keyhole interaction,heat transfer,complex melt flow,keyhole oscillation,phase transformation and microstructure evolution.Additionally,the microstructure and properties of electron beam welded joint of 2219-T6 aluminum alloy were predicted.