Research On Welding And Additive Manufacturing Process Optimization Platform For Typical Aerospace Components

In recent years,with the maturity of computer simulation technology,researchers have used computer simulation to replace expensive physical experiments,greatly improving the performance of products and the economy of preliminary design experiments.However,depending on the requirements of the project,the calculation and simulation time for some large components can take several hours,days,or even weeks.If it fails to meet the design requirements,it is necessary to repeat the trial and error iteration,which is difficult to meet the rapid response,high efficiency,and high-quality manufacturing requirements in the design and manufacturing process.The digital and intelligent level of engineering process design optimization needs to be improved.The thesis focuses on the requirements of welding and additive manufacturing hot processing technology.Taking complex structural parts as the application object,based on the adaptive surrogate model algorithm of multi-fidelity sample points,combined with the secondary development of finite element analysis software,the hot processing process simulation platform and optimization platform are constructed to realize the multifunction of additive manufacturing and welding hot processing data selection,process simulation analysis and iterative optimization,greatly shorten the hot processing process design and production preparation cycle,improve the design efficiency and manufacturing level of equipment hot processing technology,and realize the rapid and low-cost development of complex components.The main study of this thesis can be summarised as follows:(1)Analysis of the heat transfer process and heat sources of welding and additive manufacturing processes,the use of finite element simulation software to establish a finite element simulation model for multi-weld laser welding of a combustor case and SLM laser additive manufacturing of a complex cabin,respectively,and the analysis of its temperature and stress fields,and finally the prediction of the deformation of the part.(2)Construction of a multi-fidelity adaptive surrogate model for thermal processing processes.Proposed the L-H transition criterion,in the process of constructing the adaptive surrogate model,the low-precision sample points are used to explore the unknown field,and then gradually transition to the high-precision sample points to correct it.Through the comparative analysis of two examples,the exploration capability of the L-H transition criterion is verified.The comparison verifies that more low-precision sample points and fewer high-precision sample points can be added to the AMEI algorithm to construct a surrogate model with the same accuracy level,indicating the applicability of the L-H transition criterion in constructing the multi-fidelity surrogate model.The engineering cases of welding and additive manufacturing are analyzed and the surrogate model is optimized,and the constraint condition of melt pool quality is constructed for additive manufacturing.For welded parts,the deformation is optimized by 38.3%;for additive manufacturing parts,the deformation is optimized by 46.37%.The results show that the proposed algorithm can effectively optimize the process parameters of welding and additive manufacturing.(3)The secondary development of finite element software and the construction of an intelligent simulation optimization platform for hot processing technology is built.The platform is developed based on MATLAB,which can realize the functions of docking with a database,completion of experimental design,secondary development simulation call,and intelligent optimization of process parameters.