Search

This study proposes an image composition technique based on convolutional neural networks (CNNs) to construct a surrogate model for predicting fan plots of three-dimensional (3D) composite blades, which represent natural frequency lists at different rotational speeds. The proposed method composes critical 2D cross-section images to improve the accuracy of the model. Numerical examples with various compositions of cross-section images are presented to demonstrate the efficacy of the CNN model. Additionally, gradient-weighted class activation mapping analysis is used to reveal the relationship between the internal structure of the blade and the fan plots. The study shows that using multiple images in the image composition technique improves the accuracy of the model compared to using single or fewer images. Overall, the proposed method provides a promising approach for predicting fan plots of 3D composite blades using CNN models.Graphical Abstract

This paper demonstrates the possible nondestructive analysis of iron artifacts’ metallurgical characteristics using neutron imaging. Ancient kingdoms of the Korean Peninsula used a direct smelting process for ore smelting and iron bloom production; however, the use of iron blooms was difficult because of their low strength and purity. For reinforcement, iron ingots were produced through refining and forge welding, which then underwent various processes to create different iron goods. To demonstrate the potential analysis using neutron imaging, while ensuring artifacts’ safety, a sand iron ingot (SI–I) produced using ancient traditional iron making techniques and a sand iron knife (SI–K) made of SI–I were selected. SI–I was cut into 9 cm2, whereas the entirety of SI–K was preserved for analysis. SI–I was found to have an average grain size of 3 μm, with observed α-Fe (ferrite) and pearlite with a body-centered cubic (BCC) lattice structure. SI–K had a grain size of 1–3 μm, α-Ferrite on its backside, and martensite with a body-centered tetragonal (BCT) structure on its blade. Results show that the sample’s metallurgical characteristics can be identified through neutron imaging only, without losing any part of the valuable artifacts, indicating applicability to cultural artifacts requiring complete preservation.