Bull. Mater. Sci., Vol. , No. , , pp. 1--9 © Indian Academy of Sciences DOI XX.XXXX/XXXXXX-XXX-XXX-X

Predictive Modeling of Hardness Values and Phase Fraction Percentages in Micro-Alloyed Steel During Heat Treatment Using AI

wbooth school of engineering, McMaster University

MS received ; accepted

Abstract. In this work, we have proposed a predictive model that can determine the hardness and phase fraction percentages of micro-alloyed steel with a predefined chemical composition during heat treatment under specific cooling conditions. The model uses a feed-forward neural network enhanced by the ensemble method. The model has been trained on experimental data derived from Continuous Cooling Transformation (CCT) diagrams for 39 unique steel types. The inputs to the model include a cooling profile defined by a set of time-temperature values, and the chemical composition of the steel. Sensitivity analysis was performed using the validated model to understand the impact of key input variables, including individual alloys and process steps. This analysis, which measures the variability in output in response to changes in a specific input variable, showed a significant correlation with experimental findings and literature reports. Thus, our model not only predicts steel properties under varied cooling conditions but also aligns with existing theoretical knowledge and experimental data.

Keywords. Multiphase steels; Thermomechanical processing; neural network model; mechanical properties; contin- uous cooling transformation

Please follow the pdf for full paper. Kindly note that the paper is under review and subjected modification. The final published version will be shared soon