The potential use of artificial neural networks to determine the solidus temperature for steel based on composition has been investigated. Input data consist of solidus composition and temperatures both from literature and both from measurements. The primary performance testing of the model was then performed for steel grades measured. Several types of network topologies have been designed and trained and the best model selected. Testing was done on previously unseen data measured by differential thermal analysis method as on new input data. The used method is described. Obtained results are then compared to those measured and calculated by commonly used software among the academic and commercial community like IDS and Thermo-Calc. Performance of these three modelling approaches is discussed by means of selected statistic tools.Keywords: Steel, solidus temperature, artificial neural networks, Matlab, DTA
© This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.