Abstract: Producing machine learning models to capture and comprehend the relationship between variables and mechanical properties of hot-rolled steels was the goal of this effort. Mechanical Property are impacted by a variety of variables throughout the process. The innovation that has been presented would significantly alter this process. Metrics of accuracy that are utilised in the process of evaluating the effectiveness of machine learning models. To deal with complexity and uncertainty, a number of machine learning models have been created for the ultimate tensile strength, yield strength, and elongation as functions of chemical elements and thermo-mechanical variables. Machine learning techniques such as multiple linear regression, random forest model, gradient boosting model and XGBoost are used to predict mechanical properties of hot rolled steel by specifying processing parameters such as chemical composition and various thermo mechanical variables. By changing one variable while holding the other variables constant, the models were utilised to interpret trends. Spearheaded a high-impact initiative at JSW Steel to create a cutting-edge property prediction model for their hot strip mill, enhancing operational efficiency and product quality.
Abstract: Producing machine learning models to capture and comprehend the relationship between variables and mechanical properties of hot-rolled steels was the goal of this effort. Mechanical Property are impacted by a variety of variables throughout the process. The innovation that has been presented would significantly alter this process. Metrics of accura...Show More