文章信息/Info

Title:

Machine Learning Assisted High-Throughput Experiments Accelerates the Composition Design of Hard High-Entropy Alloy CoxCryTizMouWv

文章編號:

1674-3962(2020)04-0269-09

作者:

王炯1; 肖斌2; 劉軼1; 2

（1. 上海大學 材料基因組工程研究院，上海 200444）（2. 上海大學物理系 量子與分子結構國際中心，上海 200444）

Author(s):

WANG Jiong 1;  XIAO Bin 2 ;  and LIU Yi1; 2

(1. Materials Genome Institute, Shanghai University, Shanghai 200444, China) (2. International Centre for Quantum and Molecular Structures, Department of Physics, Shanghai University, Shanghai 200444, China)

關鍵詞:

高通量實驗; 機器學習; 高熵合金; 硬度

Keywords:

High-throughput experiment;  Machine learning;  High entropy alloy;  Hardness

分類號:

TP181；TG146

DOI:

10.7502/j.issn.1674-3962.201905032

文獻標志碼:

A

摘要:

針對目標性能的多元合金成分設計因具有巨大的成分參數空間而極具挑戰，而且傳統的試錯實驗由于效率低能探索的合金成分有限。提出利用高通量實驗結合機器學習方法加速非等摩爾比的硬質高熵合金CoxCryTizMouWv的成分設計。首先通過自主研發的全流程高通量合金制備系統制備了138個不同成分的高熵合金鑄態樣品。然后根據測量的維氏硬度(HV)數據，使用隨機森林法和支持向量機法進行機器學習建模，并預測了五元合金體系內潛在的3876個不同成分合金的硬度。隨機森林機器學習模型的預測結果在高（HV>800 MPa）、中（600

Abstract:

The composition design of multi-component alloy for the target performance is extremely challenging due to the enormous potential composition. The traditional trial-and�瞖rror experiments can only explore limited alloy compositions because of its low efficiency. In this work, the composition design of non-equimolar hard high-entropy alloy CoxCryTizMouWv was accelerated via combining the high-throughput experiment with machine learning. Firstly, 138 as-cast high-entropy alloys were prepared by a home-developed all-process high-throughput alloy synthesis system. Then, the machine learning models were built based on the measured Vickers hardness (HV) by using random forest (RF) and supporting vector machine methods. And, they made the prediction of HV values for 3876 potential alloys in the five�瞔omponent alloy system. The HV values predicted by RF machine learning models have the averaged errors of 2.87%, 3.30% and 6.70%, respectively in high (HV>800 MPa), medium (600