The tempering of low-alloy steels is important for controlling the mechanical properties required for industrial fields. Several studies have investigated the relationships between the input and target values of materials using machine learning algorithms. The limitation of machine learning algorithms is that the mechanism of how the input values affect the output has yet to be confirmed despite numerous case studies. To address this issue, we trained four machine learning algorithms to control the hardness of low-alloy steels under various tempering conditions. The models were trained using the tempering temperature, holding time, and composition of the alloy as the inputs. The input data were drawn from a database of more than 1900 experimental datasets for low-alloy steels created from the relevant literature. We selected the random forest regression (RFR) model to analyze its mechanism and the importance of the input values using Shapley additive explanations (SHAP). The prediction accuracy of the RFR for the tempered martensite hardness was better than that of the empirical equation. The tempering temperature is the most important feature for controlling the hardness, followed by the C content, the holding time, and the Cr, Si, Mn, Mo, and Ni contents.

For the application of Ti-6Al-4V alloys in urban air mobility, safety is very important, so achieving excellent strength and toughness is essential to prevent fractures. Regarding toughness, which is a combination of strength and ductility, it is necessary to derive the optimal heat treatment conditions for this combination of Ti-6Al-4V alloy and further understand its microstructure and fracture characteristics. For this purpose, this study investigated the microstructure in terms of grain size, plate thickness, and element distribution, as well as mechanical properties, including phase hardness and tensile properties, of Ti-6Al-4V alloy subjected to solution treatment and aging (STA) heat treatment under various aging conditions. As a result, this study suggests that solution treatment followed by aging at 630 °C for 480 min can achieve approximately 26% higher toughness than the just-solution treatment process. This is because there is little difference in hardness between the equiaxed α and basketweave structures, and β plates, which contain an excessive V between α plates, function like fibers and delay fracture.

This paper presents an experimental study of the influence of a polyaromatic additive (pyrene) on the propagation of positive streamers in cyclohexane. Experiments were carried out in gaps up to 5 cm over a wide voltage from the inception of streamers up large overvoltage. With pyrene, streamers are composed with a much larger number of branches than in pure cyclohexane. It is observed that the additive leads to slightly lower breakdown voltages, but also induces a large increase of the acceleration voltage (inception threshold of fast streamers). These results are explained by the macroscopic electrostatic properties of streamers, that depend on the streamer shape, and thus on their tendency to form branches. The transition to fast propagation modes occurs at higher voltage when streamers are more ramified.