Modeling study of the magnetocaloric behavior of La₀.₆₇Ca₀.₃₃Mn₀.₉₈Ni₀.₀₂O₃ alloys using artificial neural networks

The growing interest in environmentally friendly cooling technologies has intensified the search for efficient magnetocaloric materials (MCMs). In this context, accurately predicting the magnetocaloric effect is essential for accelerating materials discovery and reducing experimental efforts. In thi...

Full description

Saved in:
Bibliographic Details
Main Authors: Kawther Laajimi, Emna Ammar Elhadjamor, Radhouane Laajimi, Mehdi Rahmani, Mohamed Hichem Gazzah
Format: Article
Language:English
Published: Elsevier 2025-06-01
Series:Results in Engineering
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2590123025011338
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The growing interest in environmentally friendly cooling technologies has intensified the search for efficient magnetocaloric materials (MCMs). In this context, accurately predicting the magnetocaloric effect is essential for accelerating materials discovery and reducing experimental efforts. In this study, we introduce novel approaches for modeling magnetocaloric materials using artificial neural networks (ANNs). Our method aims to establish a standardized framework to evaluate the magnetocaloric effect of polycrystalline La₀.₆₇Ca₀.₃₃Mn₀.₉₈Ni₀.₀₂O₃ alloys obtained by sol-gel method. A multilayer perceptron regression model was used to predict entropy values with temperature and magnetic field variations. The model showed outstanding performance with a coefficient of determination (R²) of 0.999241 and a root mean square error (MSE) of 0.003472. In addition, the variation of heat capacity (ΔCp) under different magnetic fields was successfully predicted using the same approach, yielding a low median absolute error (MedAE) of 0.008. Our results demonstrate the potential of ANNs model to accurately approximate magnetic experimental data and highlight their ability to significantly streamline the design and optimization of newly developed MCMs.
ISSN:2590-1230