Development of novel functional concrete materials using materials informatics

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K18759
• Research Category: Grant-in-Aid for Challenging Research (Exploratory)
• Allocation Type: Multi-year Fund
• Review Section: Medium-sized Section 23:Architecture, building engineering, and related fields
• Research Institution: Tohoku University
• Principal Investigator: Nishiwaki Tomoya 東北大学, 工学研究科, 准教授 (60400529)
• Project Period (FY): 2021-07-09 – 2024-03-31
• Keywords: コンクリート / マテリアルズ・インフォマティクス / 機械学習 / 機能性コンクリート / 低炭素型コンクリート

Outline of Final Research Achievements

This research project proposed using Materials Informatics (MI), which has been remarkably developed in recent years for concrete materials, especially with new functions. The applicability of MI to material development was investigated. In particular, we focused on low-carbon concrete materials with reduced cement use. We investigated mix proportion design using compressive strength as the objective function or compressive strength prediction from mix proportions. As a result, data sets with approximately 800 data were constructed to show the possibility of predicting the relationship between mix proportion and compressive strength through machine learning. In particular, it was shown that even when multiple types of admixtures are combined, candidate mix proportions and strength prediction can be selected without the need for exhaustive experiments.

Free Research Field

建築材料学

Academic Significance and Societal Importance of the Research Achievements

世界中で広く用いられるコンクリート材料は地産地消材料の側面も持ち、幅広い原材料を組み合わせて適切な性能を得るためには、経験則による判断が行われる場合も多くある。この一方で、近年は経験則の範囲を大きく逸脱する機能性コンクリートも数多く開発されており、従来通りの実験による試行錯誤を繰り返す開発方法には効率化が求められている。本研究課題では、機械学習などを活用して新たな機能を持つコンクリートの調合設計を行う可能性を示したものである。