COMPARISON OF LINEAR REGRESSION MODELS OF THERMOPHYSICAL PROPERTIES WITH MODELS BASED ON MACHINE LEARNING

1 MACHŮ Mario
Co-authors:
1 DROZDOVÁ Ľubomíra 1 SMETANA Bedřich
Institution:
1 VSB - Technical University of Ostrava, Ostrava, Czech Republic, EU, mario.machu@vsb.cz
Conference:
31st International Conference on Metallurgy and Materials, Orea Congress Hotel Brno, Czech Republic, EU, May 18 - 19, 2022
Proceedings:
Proceedings 31st International Conference on Metallurgy and Materials
Pages:
128-133
ISBN:
978-80-88365-06-8
ISSN:
2694-9296
Published:
1st November 2022
Proceedings of the conference were published in Web of Science and Scopus.
Metrics:
580 views / 400 downloads
Abstract

The paper compares classical models for determining the thermophysical properties of steels based primarily on empirical equations derived using linear regression methods with models created using machine learning methods. The selected investigated quantities include phase transformation temperatures, specific heat capacity, coefficient of thermal expansion. The results of both approaches are verified on the measured data by methods of thermal analysis such as differential scanning calorimetry, differential thermal analysis and dilatometry. The methods are evaluated both in terms of the accuracy of predictions and in terms of the adequacy of use for a specific purpose, or in terms of the complexity of creating and using the model.

Keywords: Metallurgy, steel, properties, applications, testing methods

© 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.

Scroll to Top