Abstract: In recent years, related work on predicting the formation phase of high-entropy alloys by machine learning technology has been reported successively, but there are some disadvantages such as less empirical parameters about the formation phases of high-entropy alloys and not considering the influence of preparation process on the formation phases of high-entropy alloys. In response to this shortcoming, in this paper, 19 kinds of empirical parameters were collected based on four machine learning algorithms, namely artificial neural networks, K-nearest neighbors, support vector machines, and ensemble learning, and the 982 high-entropy alloys collected were predicted by the model. The research shows that compared with the five empirical parameters used by predecessors, the prediction accuracy of the machine learning model using 17 empirical parameters has increased from 75.75% to 79.78%. It is also found that the accuracy of the model is further improved by 4.36% compared with that of the high-entropy alloy dataset prepared by a variety of fabrication processes. The results show that adding appropriate empirical parameters and the high entropy alloy data set prepared by single casting method can improve the prediction accuracy of machine learning model.
1 Yeh J W, Chen S K, Lin S J, et al.Advanced Engineering Materials, 2004, 6, 299. 2 Cantor B, Chang I T H, Knight P, et al.Materials Science & Engineering A, 2004, 375-377,213. 3 Gurao N P, Biswas K, et al.Journal of Alloys & Compounds, 2017,697, 434. 4 Coury F G, Wilson P, Clarke K D, et al.Acta Materialia, 2019,167, 1. 5 Gao M C, Yeh J W, Liaw P K, et al.Springer International Publishing, 2016, 12,399. 6 Zhang Y, Zhou Y, Lin J, et al. Advanced Engineering Materials, 2008, 10(6),534. 7 Huang W, Martin P, Zhuang H L.Acta Materialia, 2019, 169,225. 8 Freund Y, Schapire R E.Journal of Computer & System Sciences, 1997, 55,119. 9 Wen C, Zhang Y, Wang C, et al.Acta Materialia, 2019, 170,109. 10 Roy A, Babuska T, Krick B, et al.Scripta Materialia, 2020, 185,152. 11 Dai Dongbo, Xu Tao, Wei Xiao, et al.Computational Materials Science, 2020, 175,109. 12 Zhou Z, Zhou Y, He Q, et al.npj Computational Materials, 2019, 5,128. 13 Agarwal A, Prasada R A K.JOM, 2019, 71,3424. 14 Islam N, Huang W, Zhuang H, et al.Computational Materials Science, 2018, 150,230. 15 Yeh J W, Lin S J, Chin T S, et al.Metallurgical & Materials Transactions A, 2004, 35,2533. 16 Toda-Caraballo I, Rivera-Díaz-del-Castillo P E.Intermetallics, 2016, 71,76. 17 Miracle D B, Senkov O N. Acta Materialia, 2017, 122,448. 18 Yang X, Zhang Y.Materials Chemistry and Physics, 2012, 132,233. 19 Ye Y F, Wang Q, Lu J, et al.Materials Today, 2015, 19,349. 20 Lederer Y, Toher C, Vecchio K S, et al.Acta Materialia, 2018, 159,364. 21 Gao M C, Zhang C, Gao P, et al.Current Opinion in Solid State and Materials Science, 2017, 21,238. 22 Chauhan P, Chopra S, Shanmugasundaram T.Advanced Engineering Materials, 2019, 21(1),190. 23 Zheng M, Ding W, Cao W, et al.Materials & Design, 2019, 179,78. 24 Ye Y F, Wang Q, Lu J, et al.Scripta Materialia, 2015, 104,53. 25 Singh A K, Kumar N, Dwivedi A, et al.Intermetallics, 2014, 53,112. 26 Wang Z, Qiu W, Yang Y, et al.Intermetallics, 2015, 64,63. 27 Ye Y F, Liu C T, Yang Y.Acta Materialia, 2015, 94,152. 28 He Q F, Ye Y F, Yang Y.Journal of Phase Equilibria & Diffusion, 2017, 38,416. 29 Takeuchi A, Inoue A.Materials Transactions Jim, 2007, 41(11),1372. 30 Wang Z, Huang Y, Yang Y, et al.Scripta Materialia, 2015, 94,28. 31 Melnick A B, Soolshenko V K, et al.Journal of Alloys & Compounds, 2017, 694,223. 32 Morinaga M, Yukawa H.Bulletin of Materials Science, 1997, 20,805. 33 张良均.Python数据分析与挖掘实战.机械工业出版社,2016. 34 金林,李研.统计与信息论坛,2019,34(4),3. 35 陈永星,朱胜,王晓明, 等.材料工程,2017,45(11),129. 36 李航.统计学习方法. 清华大学出版社, 2012.