MSE 2024
Lecture
24.09.2024 (CEST)
Machine learning-based magnetic property prediction for hard magnetic 14:2:1 phases
GS

Prof. Dr. rer. nat. Gerhard Schneider

Hochschule Aalen

Schneider, G. (Speaker)¹; Bernthaler, T.¹; Choudhary, A.K.¹; Goll, D.¹; Hohs, D.¹; Jansche, A.¹; Varaganti, R.¹
¹Aalen University
Vorschau
24 Min. Untertitel (CC)

Today's strongest permanent magnets, the FeNdB magnets owe their excellent magnetic properties to the intrinsic properties of the 14:2:1 phase in addition to their characteristic microstructure. The phase Fe14Nd2B has a saturation polarization of 1.61 T, a magnetocrystalline anisotropy constant of 4.3 MJ/m3 and a Curie temperature of 312 °C. Other elements, which can substitute the elements Fe, Nd or B to a certain extent, modify these intrinsic magnetic properties. We have developed regression models to predict the density, the Curie temperature as well as the saturation magnetization and the anisotropy constant at room temperature of 14:2:1 phases from the chemical composition as input features. This approach can serve as a basis for developing 14:2:1 phases with optimized magnetic properties from chemical composition.

(1) A.K. Choudhary, A. Kini, D. Hohs, A. Jansche, T. Bernthaler, O. Csiszár, D. Goll, G. Schneider, Machine learning-based Curie temperature prediction for magnetic 14:2:1 phases, AIP Advances 13, 035112 (2023); doi: 10.1063/5.0116650.

(2) A.K. Choudhary, D. Hohs, A. Jansche, T. Bernthaler, D. Goll, G. Schneider, A data-driven approach to predict the saturation magnetization for magnetic 14:2:1 phases from chemical composition, AIP Advances 14, 015060 (2024); doi: 10.1063/5.0171922.


Abstract

Abstract

Erwerben Sie einen Zugang, um dieses Dokument anzusehen.

Ähnliche Inhalte

© 2026