Numerical Simulation Of SLM Forming Process And Research And Prediction Of Forming Properties

Selective Laser Melting is an Additive Manufacturing technology suitable for metal and ceramics.Compared with the traditional manufacturing methods,the Selective Laser Melting technology has been widely utilized in many fields due to the advantages like high forming speed and high forming freedom,and so on.However,complex physical and chemical phenomena occur in the rapid forming process of Selective Laser Melting and affect the property and microstructure,and morphology of the forming components.It is challenging to reveal the impact of complex phenomena on the forming process through experiments,and it is also expensive and inefficient to adjust the process parameters to improve the performance of the formed component through the experiments.To investigate the influence of process parameters on the forming process and forming components via Selective Laser Melting in this dissertation,the melt pool dynamic model of Selective Laser Melting in the single powder layer was constructed at the mesoscopic scale.The evolution of the melt pool,the influence of forming parameters on the forming process,and the defect forming mechanism in the forming process were studied.The simulation was verified by experiments;Based on the Neural network method and the Neural network method optimized by the Genetic Algorithm,the prediction model of the mechanical properties of the formed components was established,and the accuracy of the prediction model was verified.The main works in this dissertation are as follows:(1)The three-dimensional powder layer model of 316 L stainless steel was constructed based on the Discrete Element Method,and the numerical model of the melt pool dynamics of the forming process of Selective Laser Melting at the mesoscopic scale was established through the Computational Fluid Dynamic software.The dynamic evolution of the melt pool was studied through the numerical model.The temperature field and the flow field during powder melting and solidification were also analyzed;The formation mechanism of the balling behavior and pore defect during Selective Laser Melting are revealed;The influence of forming parameters on the size of the melt pool and defects were analyzed by simulations and experiments.(2)The effect of forming parameters on the microstructure of the formed components was studied by surface topography,and the effect of forming parameters on the mechanical properties of the formed parts was analyzed by stress-strain curve and fracture morphology of the tensile samples.(3)To control the mechanical properties of formed parts by adjusting the forming parameters,taking the forming parameters(laser power,scanning speed,and scanning spacing)as the input and the tensile strength of formed components as the output,the BP prediction model of mechanical properties of formed parts is established based on the BP neural network,and the mean error is 15.31%.The Genetic algorithm is used to reduce the error of the BP neural network prediction model.The results show that the mean prediction error of the BP neural network optimized by the Genetic Algorithm is reduced to 3.54%.The Genetic Algorithm can effectively improve the accuracy of the BP neural network prediction model.This dissertation has provided theoretical support for the research of the Selective Laser Melting forming process and application guidance for adjusting forming parameters.Figure [59] table [7] reference [109]...