Dislocation nucleation in Al single crystal at shear parallel to (111) plane: Molecular dynamics simulations and nucleation theory with artificial neural networks

Improving the mechanical strength of ternary beta titanium alloy (Ti-Ta-Sn) foams, using a bimodal microstructure

This study combines an artificial neural network (ANN) and a random search to develop a system to recommend process conditions for injection molding. Both simulation and experimental results are collected using a mixed sampling method that combines Taguchi and random sampling. The dataset consists of 3600 simulations and 476 experiments from 36 different molds. Each datum has five process and 15 geometry features as input and one weight feature as output. Hyper‐parameter tuning is conducted to find the optimal ANN model. Then, transfer learning is introduced, which allows the use of simultaneous experimental and simulation data to reduce the error. The final prediction model has a root mean‐square error of 0.846. To develop a recommender system, random search is conducted using the trained ANN forward model. As a result, the weight‐prediction model based on simulated data has a relative error (RE) of 0.73%, and the weight prediction using the transfer model has an RE of 0.662%. A user interface system is also developed, which can be used directly with the injection‐molding machine. This method enables the setting of process conditions that yield parts having weights close to the target, by considering only the geometry and target weight.

https://onlinelibrary.wiley.com/doi/pdf/10.1002/aisy.202000037

Development of Artificial Neural Network System to Recommend Process Conditions of Injection Molding for Various Geometries

A neuro-genetic machine learning framework (ANN-GA) was employed to optimize and predict the optimum preparation parameters for the precipitation synthesis of high-efficiency silver-doped manganese oxides (Ag/MnOx) for toluene total oxidation The preparation conditions of Ag/MnOx for maximum CO2 yield were predicted to be at 132 wt% of silver loading, 415 min of stirring time, 461 ℃ of calcination temperature, and 43 h of calcination time The resulting Ag/MnOx-GA catalyst achieved the lowest T50 (CO2) of 206 ℃ comparing to the catalyst suggested by the response surface methodology (RSM) (214 ℃) and synthesized catalyst Ag/MnOx-R16 (220 ℃) The obtained relative importance of each factor involved in the preparation process revealed that all of the parameters should be taken into account, among which the silver loading had the most significant impact on the catalytic reactivity The synthesized materials were characterized by ICP-OES, SEM/EDS, BET, Raman, XRD, XPS, TGA, H2-TPR, and O2-TPD The results indicated that the Ag/MnOx-GA exhibited the largest surface area (844), the lowest average oxidation state (AOS) of Mn (325), the best redox capacity, and the highest ratios of Oads/Olatt (075) These consequences provided evidence that the toluene total oxidation was more likely to occur on the surface of Ag/MnOx-GA at a relatively lower temperature

Neuro-genetic machine learning framework accelerates the optimization of Ag/MnOx catalyst for total oxidation of toluene

Transfer learning aid the prediction of sintering densification

Analysis of cold compaction for Fe-C, Fe-C-Cu powder design based on constitutive relation and artificial neural networks

Changing the material parameters such as powder characteristics and additives affects the final properties of an iron–carbon alloy. This study investigated the influences of three typical material parameters, iron particle size, graphite addition, and powder lubricant addition, on the density and mechanical properties of an iron–carbon alloy formed via powder compaction and sintering. Each material parameter was designed with five levels, and all of the powder mixtures were compacted under 500 MPa and sintered at 1120 °C for 30 min. The microstructure of the samples was observed for the green part and sintered part. Through the tensile test, yield strength, ultimate tensile strength, and elongation were measured. The tensile fracture surface was also examined to understand the changes in mechanical properties according to the parameters. The correlations between mechanical properties and material parameters were characterized by the mapping functions, and a sensitivity analysis was carried out to investigate which parameter had the larger influence on the mechanical properties. The results showed that graphite addition has the greatest influence on the mechanical properties due to the microstructural changes from hypoeutectoid structure to hypereutectoid structure. Further, a regression model was developed to describe the mechanical response of the iron–carbon alloy depending on the material conditions.

Suk Hyun Kim

Papers

Analysis of cold compaction for Fe-C, Fe-C-Cu powder design based on constitutive relation and artificial neural networks

Correlation Study Between Material Parameters and Mechanical Properties of Iron–Carbon Compacts Using Sensitivity Analysis and Regression Model