PEAK POINT DESCRIPTION UTILIZING OF AN ARTIFICIAL NEURAL NETWORK APPROACH IN COMPARISON WITH THE COMMONLY USED RELATIONSHIPS

1 Opěla Petr
Co-authors:
1 Schindler Ivo 1 Rusz Stanislav 1 Ševčák Vojtěch
Institution:
1 VSB – Technical University of Ostrava, Ostrava, Czech Republic, EU, petr.opela@vsb.cz
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:
432-437
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:
23 views / 6 downloads
Abstract

The peak point coordinates (i.e. peak stress, peak strain) play a significant role in case of a flow curve description. These coordinates are strongly dependent on the temperature and strain rate, so they need to be related to these thermomechanical circumstances before use in the flow stress models. In this research, the experimental peak point coordinates of the C45 and 38MnVS6 steels were described in a wide range of thermomechanical conditions by use of two different methodologies. The first one was based on the ordinary predictive relationships utilizing the well-known Zener-Hollomon parameter. The second one was based on the artificial neural network approach. The aim was to compare appropriateness of these methods. The results have suggested better aptness in case of the assembled neural networks.

Keywords: Peak point coordinates, predictive relationship, artificial neural network
Scroll to Top