| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 1562024 | Computational Materials Science | 2011 | 8 Pages |
Artificial neural networks have been used to estimate the volume fraction of bainite in low carbon steels containing various alloying elements. The network predicts the volume fraction for a given composition, isothermal transformation temperature and isothermal transformation time. Additionally, the maximum transformation temperature at which bainite formation takes place is also provided as an input to the neural network. The network was trained using the experimental data from three low carbon steels and it was found to perform quite well in predicting the volume fraction of bainite. The impact of the composition of alloying elements on the volume fraction of bainite was also studied and the results were in agreement with the known metallurgical theory.
► Neural network applied to determine volume fraction of bainite. ► Effect of alloying elements content on volume fraction studied. ► Neural network predictions compared with metallurgical theory.
