Artificial intelligence-assisted non-metallic inclusion particles analysis in advanced steels using machine learning: a review

Detection and characterization of non-metallic inclusions are of vital importance for clean steel production. Recently, the imaging analysis in combination of high-dimensional data processing of metallic materials according to artificial intelligence (AI)-based machine learning (ML) method has developed rapidly. This technique achieves impressive results in the field of inclusions classification in process metallurgy. The current work surveys the ML modelling of inclusion prediction in advanced steels including the detection, classification and feature prediction of inclusions in different steel grades. Data analysis and image analysis of clean steel studies with different features by ML are summarized. In terms of data analysis, the inclusion prediction methodology based on ML establishes a connection between experimental parameters and inclusion characteristics and analyzes the experimental parameters importance. In terms of image analysis, the focus is placed on the classification of different types of inclusions with deep learning (DL) method in comparison with data analysis. Finally, the further development of inclusion analysis by ML-based methods is recommended. This work paves the way for the application of AI-based methodology for ultra-clean steel study towards a sustainable metallurgy perspective.