Development and experimental validation of a neural network model for prediction and analysis of the strength of bainitic steels

In this investigation, a neural network model is developed to predict the hardness of high carbon steels. The inputs to the neural network include the weight percentage of nine alloying elements and the heat treatment conditions such as austenitization temperature and isothermal transformation temperature and time. For the model development, 15 steels from the literature were used. The developed model was validated with respect to eight other steels from the literature that were not used for the model development. Additionally, the model was also employed to predict the hardness of five newly designed bainitic steels. Further, from the new experimental data, identifying the steel containing Co and Al as potentially viable for mass production, a computational analysis of this steel using the developed neural network model indicates the possibility of minimizing the processing costs by adjusting the alloying element content, Co in particular.

► Twenty seven new heat treatment experiments on five new nanobainitic steels. ► Modeling for hardness of steel, applicable for process control in industry. ► Extensive validation of the model for steel industrial implementation. ► Comparison with recent hardness models.