Finite Element Analysis and Computational Fluid Dynamics Verification of Molten Pool Characteristics During Selective Laser Melting of Ti-6Al-4V Plates.

The finite element (FE) method is used to characterize the thermal gradient, solidification rate, and molten pool sizes of Ti-6Al-4V plates in the process of selective laser melting (SLM). The results are verified by using the computational fluid dynamics (CFD) simulation. The proposed FE model contains a series of toolpath information that is directly converted from a G-code file, including hatch spacing, laser power, layer thickness, dwell time, and scanning speed generated by using Slic3r software from a CAD file. A proposed multi-layer, multi-track FE model is used to investigate the influence of the laser power, scanning speed, and scanning path on the microstructure in the Ti-6Al-4V plate built via SLM. The processing window is also determined based on the proposed FE model. The FE results indicate that, with a decrease in the laser power and an increase in the scanning speed, the morphology of the crystal grains, showing fully columnar crystals, gradually deviates from the fully equiaxed region. The formed grains are dependent on the laser power, scanning speed, and deposition position, but they are not sensitive to the scanning path, and with the deposition from the bottom layer to the top layer, the size of the formed grains is gradually increasing, which shows a good agreement with the experimental results.

Optimized XGBoost Model with Small Dataset for Predicting Relative Density of Ti-6Al-4V Parts Manufactured by Selective Laser Melting.

Determining the quality of Ti-6Al-4V parts fabricated by selective laser melting (SLM) remains a challenge due to the high cost of SLM and the need for expertise in processes and materials. In order to understand the correspondence of the relative density of SLMed Ti-6Al-4V parts with process parameters, an optimized extreme gradient boosting (XGBoost) decision tree model was developed in the present paper using hyperparameter optimization with the GridsearchCV method. In particular, the effect of the size of the dataset for model training and testing on model prediction accuracy was examined. The results show that with the reduction in dataset size, the prediction accuracy of the proposed model decreases, but the overall accuracy can be maintained within a relatively high accuracy range, showing good agreement with the experimental results. Based on a small dataset, the prediction accuracy of the optimized XGBoost model was also compared with that of artificial neural network (ANN) and support vector regression (SVR) models, and it was found that the optimized XGBoost model has better evaluation indicators such as mean absolute error, root mean square error, and the coefficient of determination. In addition, the optimized XGBoost model can be easily extended to the prediction of mechanical properties of more metal materials manufactured by SLM processes.

Oscillators based on double-walled armchair@zigzag carbon nanotubes containing inner tubes with different helical rises.

A novel approach is presented to improve the oscillatory behavior of oscillators based on double-walled carbon nanotubes containing rotating inner tubes applied with different helical rises. The influence of the helical rise on the oscillatory amplitude, frequency, and stability of inner tubes with different helical rises in armchair@zigzag bitubes is investigated using the molecular dynamics method. Our simulated results show that the oscillatory behavior is very sensitive to the applied helical rise. The inner tube with h = 10 Å has the most ideal hexagon after the energy minimization and NVT process in the armchair@zigzag bitubes, superior even to the inner tube without a helical rise, and thus it exhibits better oscillatory behavior compared with other modes. Therefore, we can apply an appropriate helical rise on the inner tube to produce a stable and smooth oscillator based on double-walled carbon nanotubes.

Homogenized finite element analysis on effective elastoplastic mechanical behaviors of composite with imperfect interfaces.

A three-dimensional (3D) representative volume element (RVE) model was developed for analyzing effective mechanical behavior of fiber-reinforced ceramic matrix composites with imperfect interfaces. In the model, the fiber is assumed to be perfectly elastic until its tensile strength, and the ceramic material is modeled by an elasto-plastic Drucker-Prager constitutive law. The RVE model is then used to study the elastic properties and the tensile strength of composites with imperfect interfaces and validated through experiments. The imperfect interfaces between the fiber and the matrix are taken into account by introducing some cohesive contact surfaces. The influences of the interface on the elastic constants and the tensile strengths are examined through these interface models.

Steered molecular dynamics simulations on the peeling and shearing of carbon nanotubes on a silicon substrate.

Steered molecular dynamics simulations were performed to investigate the peeling and shearing behavior of a single-walled carbon nanotube lying on a silicon substrate. Both the constant velocity and the constant force methods were applied to explore the adsorption of carbon nanotube and silicon substrate, and the efficiency of the two simulation methods was compared via a few representative examples. We examined the influences of the peeling angle, the shearing velocity, the initial distance between the carbon nanotube and the substrate, the connection point with the virtual ideal spring, the tube radius, as well as the 5-7-7-5 and radius defect of the carbon nanotube. The numerical results coincide well with relevant experimental results. This work is helpful for the application of carbon nanotubes in silicon-based microelectronics.