[1]王炯,肖斌,劉軼.機(jī)器學(xué)習(xí)輔助的高通量實(shí)驗(yàn)加速硬質(zhì)高熵合金 CoxCryTizMouWv成分設(shè)計(jì)[J].中國材料進(jìn)展,2020,(04):269-277.[doi:10.7502/j.issn.1674-3962.201905032]
WANG Jiong,XIAO Bin,and LIU Yi.Machine Learning Assisted High-Throughput Experiments Accelerates the Composition Design of Hard High-Entropy Alloy CoxCryTizMouWv[J].MATERIALS CHINA,2020,(04):269-277.[doi:10.7502/j.issn.1674-3962.201905032]
點(diǎn)擊復(fù)制
機(jī)器學(xué)習(xí)輔助的高通量實(shí)驗(yàn)加速硬質(zhì)高熵合金 CoxCryTizMouWv成分設(shè)計(jì)(
)
中國材料進(jìn)展[ISSN:1674-3962/CN:61-1473/TG]
- 卷:
-
- 期數(shù):
-
2020年第04期
- 頁碼:
-
269-277
- 欄目:
-
- 出版日期:
-
2020-04-30
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(xué) 材料基因組工程研究院,上海 200444)(2. 上海大學(xué)物理系 量子與分子結(jié)構(gòu)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(guān)鍵詞:
-
高通量實(shí)驗(yàn); 機(jī)器學(xué)習(xí); 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(xiàn)標(biāo)志碼:
-
A
- 摘要:
-
針對目標(biāo)性能的多元合金成分設(shè)計(jì)因具有巨大的成分參數(shù)空間而極具挑戰(zhàn),而且傳統(tǒng)的試錯(cuò)實(shí)驗(yàn)由于效率低能探索的合金成分有限。提出利用高通量實(shí)驗(yàn)結(jié)合機(jī)器學(xué)習(xí)方法加速非等摩爾比的硬質(zhì)高熵合金CoxCryTizMouWv的成分設(shè)計(jì)。首先通過自主研發(fā)的全流程高通量合金制備系統(tǒng)制備了138個(gè)不同成分的高熵合金鑄態(tài)樣品。然后根據(jù)測量的維氏硬度(HV)數(shù)據(jù),使用隨機(jī)森林法和支持向量機(jī)法進(jìn)行機(jī)器學(xué)習(xí)建模,并預(yù)測了五元合金體系內(nèi)潛在的3876個(gè)不同成分合金的硬度。隨機(jī)森林機(jī)器學(xué)習(xí)模型的預(yù)測結(jié)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-anderror 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 fivecomponent 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
備注/Memo
- 備注/Memo:
-
收稿日期:2019-05-26 基金項(xiàng)目:國家科技部重點(diǎn)研發(fā)計(jì)劃“材料基因組工程”項(xiàng)目(2017YFB0702901,2017YFB0701502);國家自然科學(xué)基金項(xiàng)目(91641128)第一作者:王炯,男,1990年生,碩士研究生通訊作者:劉軼,男,1971年生,教授,博士生導(dǎo)師, Email:yiliu@t.shu.edu.cn
更新日期/Last Update:
2020-03-26