Machine learning enabling prediction in mechanical performance of Ti6Al4V fabricated by large-scale laser powder bed fusion via a stacking model

Determining appropriate process parameters in large-scale laser powder bed fusion(LPBF)additive manufacturing pose formidable challenges that necessitate advanced approaches to minimize trial-and-error during experimentation.This work proposed a data-driven approach based on stacking ensemble learning to predict the mechanical properties of Ti6Al4V alloy fabricated by large-scale LPBF for the first time.This method can adapt to the complexity of large-scale LPBF data distribution and exhibits a more generalized predictive capability compared to base models.Specifically,the stacking model utilized artificial neural network(ANN),gradient boosting regressor,kernel ridge regression,and elastic net as base models,with the Lasso model serving as the meta-model.Bayesian optimization and cross-validation were utilized for model optimization and training based on a limited data set,resulting in higher predictive accuracy compared to traditional artificial neural network model.The statistical analysis of the ANN and stacking models indicates that the stacking model exhibits superior performance on the test set,with a coefficient of determination value of 0.944,mean absolute percentage error of 2.51%,and root mean squared error of 27.64,surpassing that of the ANN model.All statistical metrics demonstrate superiority over those obtained from the ANN model.These results confirm that by integrating the base models,the stacking model exhibits superior predictive stability compared to individual base models alone,thereby providing a reliable assessment approach for predicting the mechanical properties of metal parts fabricated by the LPBF process.

Research on CGI in Embedded System

This paper mainly studies the method of dynamic embedded Web server technology and its realization. Taking S3C2440 processor as the core hardware platform, constructed the software system of based on Linux operating system on the hardware platform; Analysis the key technology of web server, select Boa as the embedded web server, Boa server and CGIC database successfully transplanted and run the static Webpage： The paper detailed analysis of the CGI technology and using C language to compile the CGI program to realize dynamic Web server, realize the use of the Web browser to the remote Web server access control function.

Influence Mechanism of Process Parameters on Relative Density, Influence Mechanism of Process Parameters on Relative Density, Microstructure, and Mechanical Properties of Low Sc-Content Al-Mg-Sc-Zr Alloy Fabricated by Selective Laser Melting

Additive manufacturing of Al-Mg-Sc-Zr alloys is a promising technique for the fabrication of lightweight components with complex shapes.In this study,the effect of the process parameters of selective laser melting(SLM)on the surface morphology,relative density,microstructure,and mechanical properties of Al-Mg-Sc-Zr high-strength aluminum alloys with low Sc content was systematically investigated.The results show that the energy density has an important effect on the surface quality and densification behavior of the Al-Mg-Sc-Zr alloy during the SLM process.As the energy density increased,the surface quality and the number of internal pores increased.However,the area of the fine-grained region at the boundary of the molten pool gradually decreased.When the laser energy density was set to 151.52 J/mm3,a low-defect sample with a relative density of 99.2%was obtained.After heat treatment,the area of the fine grains at the boundary increased significantly,thereby contributing to the excellent mechanical properties.The microstructure was characterized by a unique“fan-shaped”heterogeneous structure.As the energy density increased,the microhardness first increased and then decreased,reaching a maximum value of 122 HV0.3.With the optimized process parameters,the yield strength(YS),ultimate tensile strength(UTS),and elongation of the as-built Al-Mg-Sc-Zr alloys were 346.8±3.0 MPa,451.1±5.2 MPa,14.6%±0.8%,respectively.After heat treatment at 325°C for 8 h,the hardness increased by 38.5%to 169 HV0.3,and the YS and UTS increased by 41.3%and 18.1%,respectively,to 490.0±9.0 MPa and 532.7±7.8 MPa,respectively,while the elongation slightly decreased to 13.1%±0.7%.

Role of heterogenous microstructure and deformation behavior in achieving superior strength-ductility synergy in zinc fabricated via laser powder bed fusion

Zinc(Zn)is considered a promising biodegradable metal for implant applications due to its appropriate degradability and favorable osteogenesis properties.In this work,laser powder bed fusion(LPBF)additive manufacturing was employed to fabricate pure Zn with a heterogeneous microstructure and exceptional strength-ductility synergy.An optimized processing window of LPBF was established for printing Zn samples with relative densities greater than 99%using a laser power range of 80∼90 W and a scanning speed of 900 mm s−1.The Zn sample printed with a power of 80 W at a speed of 900 mm s−1 exhibited a hierarchical heterogeneous microstructure consisting of millimeter-scale molten pool boundaries,micrometer-scale bimodal grains,and nanometer-scale pre-existing dislocations,due to rapid cooling rates and significant thermal gradients formed in the molten pools.The printed sample exhibited the highest ductility of∼12.1%among all reported LPBF-printed pure Zn to date with appreciable ultimate tensile strength(∼128.7 MPa).Such superior strength-ductility synergy can be attributed to the presence of multiple deformation mechanisms that are primarily governed by heterogeneous deformation-induced hardening resulting from the alternative arrangement of bimodal Zn grains with pre-existing dislocations.Additionally,continuous strain hardening was facilitated through the interactions between deformation twins,grains and dislocations as strain accumulated,further contributing to the superior strength-ductility synergy.These findings provide valuable insights into the deformation behavior and mechanisms underlying exceptional mechanical properties of LPBF-printed Zn and its alloys for implant applications.