기계학습에 기반한 AA1100 합금의 부식에 미치는 환경 영향 분석

• Published in: Corrosion science and technology Vol. 24; no. 4; pp. 269 - 280
• Main Authors: 정용주, Yong Ju Jeong, 김민태, Mintae Kim, 서정우, Jeongwoo Seo, 이주희, Ju Hee Lee, 장희진, Hee Jin Jang, 김희수, Hee-soo Kim
• Format: Journal Article
• Language: Korean
• Published: 한국부식방식학회 31.08.2025
• Subjects: Aluminum alloys; Corrosion; Machine learning; Polarization curve; 금속공학
• ISSN: 1598-6462; 2288-6524
• DOI: 10.14773/cst.2025.24.4.269

This study explores the effects of three environmental chemical species―Na₂S, NaCl, and H₂O₂―on the corrosion behavior of AA1100 aluminum alloy using machine learning (ML) techniques. We collected experimental data through a full factorial design that included 27 solution conditions, from which we extracted two key electrochemical parameters: corrosion potential (E corr ) and corrosion current density (i corr ), We trained four regression algorithms―k-nearest neighbor, random forest, support vector regression, and extreme gradient boosting―on this data and compared their performance using root mean squared error. Among these models, the random forest algorithm demonstrated the most accurate predictions for both E corr and i corr , leading us to select it for further analysis. To assess the influence of each input variable, we employed SHAP (SHapley Additive exPlanations) analysis. Our findings revealed that NaCl made the greatest positive contribution to E corr , while H₂O₂ significantly increased i corr . In contrast, Na₂S had minimal impact on both corrosion indicators. SHAP interaction plots indicated that Cl⁻ and H₂O₂ had synergistic effects in accelerating corrosion, while Na₂S remained inert. This study highlights the effectiveness of interpretable ML approaches in analyzing complex corrosion systems and offers a data-driven perspective for corrosion prediction and material design.