Machine Learning Methods for the Prediction of the Inclusion Content of Clean Steel Fabricated by Electric Arc Furnace and Rolling

Machine Learning classification models have been trained and validated from a dataset (73 features 13,616 instances) including experimental information of clean cold forming steel fabricated by electric arc furnace hot rolling. A model was developed to identify inclusion contents above the median. The following algorithms were implemented: Logistic Regression, K-Nearest Neighbors, Decision Tree, Random Forests, AdaBoost, Gradient Boosting, Support Vector Classifier Artificial Neural Networks. Forest displayed best results overall selected for subsequent analyses. Permutation Importance method used variables that influence cleanliness impact these determined means Partial Dependence Plots. final diameter coil has interpreted considering changes induced process rolling in distribution inclusions. Several related secondary metallurgy tundish operations identified metallurgical terms. In addition, inspection area during microscopic examination samples also appears content. Recommendations established sampling manufacturing conditions optimize inclusionary steel.