東京大学 · 工学
本研究室では、機械学習を活用した新規多要素合金の創出を柱とし、特に一般化回帰ニューラルネットワーク（GRNN）を用いた組成-硬度予測モデルの構築により、超高硬度を示すFe-based MPEAsの設計と合成に成功しています。また、Fe-Znインタメタル化合物の界面挙動や破壊靭性のメカニズム解明、軟岩内の局所化ひずみの発展挙動の実験的・数値的解析についても進めています。材料の力学的・微視的特性の高精度な予測と制御を目指した、計算材料工学と実験材料科学の融合的アプローチが特徴です。
Figures are computed from collected data and may differ slightly.
The Fe-Zn intermetallic compounds (IMC) layers composed of δp, δk, and Γ phases were fabricated using two different kinds of Fe/Zn diffusion couple (DC), and the fracture toughness of the constituent phases was estimated from the toughness of the IMC layers. In the DCs with sufficient Zn supply, the IMC layers were mainly composed of δp phase after isothermal holding at 450-600 °C for 60 s, while in the DCs with limited Zn supply, the IMC layers were composed of δp, δk, and Γ phases at the early
Abstract Strain localization developing inside soft rock specimens is examined through experimental observation and numerical simulation. In the experimental study, soft rock specimens are sheared at different strain rates under plane strain conditions and deformation and strain localization characteristics are analysed. Transition of localization mode from highly localized mode for higher strain rate to distributed and diffused mode of strain localization for lower strain rates was observed. In
In this work, machine learning (ML) technique was used to discovery new multi-principal elements alloys (MPEAs) with desirable properties. Generalized Regression Neural Network (GRNN) showed high accuracy to construct the composition-microhardness model and was used for microhardness prediction and composition optimization. Based on ML results, Fe0.6Ni0.7CrAl MPEAs were designed and prepared. The proposed GRNN model aligns well with experimental data, Fe0.6Ni0.7CrAl MPEAs exhibit ultra-high micr
Open papers in the app to read, cite, and organize with AI.
