Composition design for (PrNd–La–Ce)<sub>2</sub>Fe<sub>14</sub>B melt-spun magnets by machine learning technique

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

Li Rui^{1, 2}, Liu Yao^{1, 2}, Zuo Shu-Lan^{1, 2}, Zhao Tong-Yun^{1, 2}, Hu Feng-Xia^{1, 2}, Sun Ji-Rong^{1, 2}, Shen Bao-Gen^{1, 2, †}

Predicted maps of (a) Hcj and (b) (BH)_max using the best models (GBRT) for (PrNd_{1-Ce_p}Ce_{Ce_P})_REsFe_93-REsTM₁B₆ melt-spun magnets, and their corresponding cost performances. Circle represents a pure Nd magnet with H_cj about 11.0 kOe, which can be replaced by the economical Ce-containing magnets with the same H_cj at the expense of (BH)_max (Star).