Author(s)

Mengru Li and Xiaoyu Zhou

Corresponding Author

Mengru Li

ABSTRACT

The magnetic refrigeration technology based on the magnetocaloric effect (MCE) has been considered as a more advantageous technique over present well-used conventional gas compression. However, the composition of a vast number of magnetic materials with large MCE are still remained unknown. On the other hand, data-mining techniques using machine learning are found really efficient in many fields, especially for materials design. So here, due to the distinctive advantages of amorphous alloy over crystalline materials and the intrinsic large MCE of heavy rare earth-based alloys, the system of rare earth-based amorphous alloy was selected to study by this technique. By using a machine learning algorithm called gradient boosting regression tree (GBRT) and putting the external magnetic field into input features together with the composition, two models were successfully built to predict Curie temperature (TC) and magnetic entropy change (∆SM) with high accuracy. The performance metric coefficient scores of determination (R2) of the two models are 0.96 and 0.91. Finally, because of the error that exists in the ∆SM itself, a new standard was presented to clearly see if the data of ∆ SM predicted by us can be accepted. The success of the two models building and their excellent generalization ability suggest that they will be really helpful for our experiments guided and find proper composition for further magnetic refrigeration applications.

KEYWORDS

Rare earth-based amorphous alloy, machine learning, magnetic refrigeration, composition design