Evaluation of Aging Grades of Heat-Resistant Steel Based on Portable LIBS

Heat-resistant steels are commonly used in high-temperature and high-pressure vessels and boilers. The microstructure and mechanical properties of heat-resistant steels degrade during service, and real-time, rapid monitoring of the aging status of heat-resistant steels is crucial for safe operation and production. Laser Induced Breakdown Spectroscopy (LIBS) is an emerging atomic emission spectroscopy technique. This study utilizes portable LIBS equipment to quickly diagnose the aging level of T91 heat-resistant steel, and optimizes the dimensionality reduction and modeling methods for the obtained spectral features. Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) were employed as dimensionality reduction methods to optimize and simplify the spectral features. Based on the reduced data, K-Nearest Neighbor (KNN) and Support Vector Machine (SVM) algorithms were further applied to establish a steel aging grade assessment model, and the impact of key parameter selection on model performance was discussed. The results show that the spectral data reduced by LDA achieve better clustering distribution, which can significantly improve the accuracy of the assessment model. Additionally, the LDA-SVM model yields the highest aging level assessment accuracy of 94.58%. The modeling methods employed in this study can effectively realize the aging level evaluation of T91 heat-resistant steel based on portable LIBS.