Application of neural networks for designing the chemical composition of steel with the assumed hardness after cooling from the austenitising temperature

The method presented in the paper makes it possible to determine the mass concentrations of the alloying elements for steels with the required curve of hardness changes versus cooling rate. Search for the optimum chemical composition is carried out in two stages. The first stage consists in preparing the data file consisting of chemical compositions of steels and calculated curves of hardness change versus cooling rate. Hardness of steel cooled from the austenitising temperature is calculated with the model using the artificial neural networks. At the second stage, the chemical composition of steel is searched for the closest to the assumed criterion.