中国物理B ›› 2018, Vol. 27 ›› Issue (4): 47501-047501.doi: 10.1088/1674-1056/27/4/047501

• CONDENSED MATTER: ELECTRONIC STRUCTURE, ELECTRICAL, MAGNETIC, AND OPTICAL PROPERTIES • 上一篇    下一篇

Composition design for (PrNd-La–Ce)2Fe14B melt-spun magnets by machine learning technique

Rui Li(李锐), Yao Liu(刘瑶), Shu-Lan Zuo(左淑兰), Tong-Yun Zhao(赵同云), Feng-Xia Hu(胡凤霞), Ji-Rong Sun(孙继荣), Bao-Gen Shen(沈保根)   

  1. 1. State Key Laboratory of Magnetism, Institute of Physics, Chinese Academy of Sciences, Beijing 100190, China;
    2. University of Chinese Academy of Sciences, Beijing 100049, China
  • 收稿日期:2018-01-14 修回日期:2018-02-27 出版日期:2018-04-05 发布日期:2018-04-05
  • 通讯作者: Bao-Gen Shen E-mail:shenbg@iphy.ac.cn
  • 基金资助:

    Project supported by the National Basic Research Program of China (Grant No. 2014CB643702), the National Natural Science Foundation of China (Grant No. 51590880), the Knowledge Innovation Project of the Chinese Academy of Sciences (Grant No. KJZD-EW-M05), and the National Key Research and Development Program of China (Grant No. 2016YFB0700903).

Composition design for (PrNd-La–Ce)2Fe14B melt-spun magnets by machine learning technique

Rui Li(李锐)1,2, Yao Liu(刘瑶)1,2, Shu-Lan Zuo(左淑兰)1,2, Tong-Yun Zhao(赵同云)1,2, Feng-Xia Hu(胡凤霞)1,2, Ji-Rong Sun(孙继荣)1,2, Bao-Gen Shen(沈保根)1,2   

  1. 1. State Key Laboratory of Magnetism, Institute of Physics, Chinese Academy of Sciences, Beijing 100190, China;
    2. University of Chinese Academy of Sciences, Beijing 100049, China
  • Received:2018-01-14 Revised:2018-02-27 Online:2018-04-05 Published:2018-04-05
  • Contact: Bao-Gen Shen E-mail:shenbg@iphy.ac.cn
  • Supported by:

    Project supported by the National Basic Research Program of China (Grant No. 2014CB643702), the National Natural Science Foundation of China (Grant No. 51590880), the Knowledge Innovation Project of the Chinese Academy of Sciences (Grant No. KJZD-EW-M05), and the National Key Research and Development Program of China (Grant No. 2016YFB0700903).

RichHTML

0

PDF (PC)

247

赞

摘要：

Data-driven technique is a powerful and efficient tool for guiding materials design, which could supply as an alternative to trial-and-error experiments. In order to accelerate composition design for low-cost rare-earth permanent magnets, an approach using composition to estimate coercivity (Hcj) and maximum magnetic energy product ((BH)max) via machine learning has been applied to (PrNd-La-Ce)2Fe14B melt-spun magnets. A set of machine learning algorithms are employed to build property prediction models, in which the algorithm of Gradient Boosted Regression Trees is the best for predicting both Hcj and (BH)max, with high accuracies of R2=0.88 and 0.89, respectively. Using the best models, predicted datasets of Hcj or (BH)max in high-dimensional composition space can be constructed. Exploring these virtual datasets could provide efficient guidance for materials design, and facilitate the composition optimization of 2:14:1 structure melt-spun magnets. Combined with magnets' cost performance, the candidate cost-effective magnets with targeted properties can also be accurately and rapidly identified. Such data analytics, which involves property prediction and composition design, is of great time-saving and economical significance for the development and application of LaCe-containing melt-spun magnets.

关键词: permanent magnet, materials design, machine learning, property prediction

Abstract:

Data-driven technique is a powerful and efficient tool for guiding materials design, which could supply as an alternative to trial-and-error experiments. In order to accelerate composition design for low-cost rare-earth permanent magnets, an approach using composition to estimate coercivity (Hcj) and maximum magnetic energy product ((BH)max) via machine learning has been applied to (PrNd-La-Ce)2Fe14B melt-spun magnets. A set of machine learning algorithms are employed to build property prediction models, in which the algorithm of Gradient Boosted Regression Trees is the best for predicting both Hcj and (BH)max, with high accuracies of R2=0.88 and 0.89, respectively. Using the best models, predicted datasets of Hcj or (BH)max in high-dimensional composition space can be constructed. Exploring these virtual datasets could provide efficient guidance for materials design, and facilitate the composition optimization of 2:14:1 structure melt-spun magnets. Combined with magnets' cost performance, the candidate cost-effective magnets with targeted properties can also be accurately and rapidly identified. Such data analytics, which involves property prediction and composition design, is of great time-saving and economical significance for the development and application of LaCe-containing melt-spun magnets.

Key words: permanent magnet, materials design, machine learning, property prediction

中图分类号:  (Metals and alloys)

  • 75.47.Np
75.50.Ww (Permanent magnets)