ESTIMATION OF LIQUIDUS TEMPERATURES OF STEEL USING ARTIFICIAL NEURAL NETWORK APPROACH

1,2 MACHŮ Mario
Co-authors:
1,2 DROZDOVÁ Ľubomíra 1,2 SMETANA Bedřich 1 ZIMNÝ Ondřej 1,2 VLČEK Jozef
Institutions:
1 VSB - Technical University of Ostrava, 17. listopadu 15, Ostrava-Poruba, 708 33, Ostrava, Czech Republic, EU, mario.machu@vsb.cz
2 Regional material science and technology centre, VŠB-Technical University of Ostrava, 17. listopadu 15, Ostrava-Poruba, 708 33, Czech Republic, EU
Conference:
27th International Conference on Metallurgy and Materials, Hotel Voronez I, Brno, Czech Republic, EU, May 23rd - 25th 2018
Proceedings:
Proceedings 27th International Conference on Metallurgy and Materials
Pages:
56-62
ISBN:
978-80-87294-84-0
ISSN:
2694-9296
Published:
24th October 2018
Proceedings of the conference were published in Web of Science and Scopus.
Metrics:
39 views / 18 downloads
Abstract

Presented works investigates a possibility of using modeling based on artificial neural network for prediction of liquidus temperatures of low-alloyed steels. Paper describes the methodology of creating such model by tools incorporated in commercial software MATLAB. Neural network is trained, validated and tested and previously unseen data measured by DTA method are used as new input data. Results are then compared to those measured and calculated by commonly used software for such applications like IDS and Thermo-Calc. Performance of these three modeling approaches is discussed.

Keywords: artificial neural networks, liquidus temperature of steel, MATLAB, Thermo-Calc, IDS
Scroll to Top