数据驱动的航空发动机材料设计研究进展  

Progress of Data-Driven Aero-Engine Materials Design

作　　者：袁睿豪 廖玮杰 唐斌[1] 樊江昆[1] 王军[1] 寇宏超[1] 李金山[1] YUAN Ruihao;LIAO Weijie;TANG Bin;FAN Jiangkun;WANG Jun;KOU Hongchao;LI Jinshan(State Key Laboratory of Solidification Processing,Northwestern Polytechnical University,Xi’an 710072,China)

机构地区：[1]西北工业大学凝固技术国家重点实验室,西安710072

出　　处：《航空制造技术》2021年第18期22-30,共9页Aeronautical Manufacturing Technology

基　　金：国家重大科技专项(MJ-2018-G-48)。

摘　　要：数据驱动的材料设计模式已经广泛应用于材料科学研究中,以加速新材料从研发到应用的进程。机器学习技术能够挖掘数据中存在的潜在模式或规律,是数据驱动材料设计模式的研究关键和热点。其中,基于主动学习的策略已经成功指导了多种新材料的开发。首先,介绍了主动学习的研究思路、构成要素及每个环节常用的方法;其次,以航空发动机材料如高温合金、钛基合金、复合材料、热障涂层等为例,介绍了机器学习在其中的具体应用和取得的效果;最后,针对航空发动机材料的恶劣服役环境,展望了机器学习应用于航空发动机材料面临的挑战,并提出了可能的解决方案。The data-driven materials design paradigm has been largely applied in the research of materials science,to shorten the process from materials discovery to applications.As the key and hot pot in the research of data-driven materials design,machine learning can extract the pattern or law behind the data,of which the active-learning based strategy has been successfully employed to guide the exploitation of materials.Firstly,this paper introduces the main idea and elements of the active learning loop,as well as the methods/algorithms used in each step.Then the paper reviews the applications and effect of machine learning in the field of aero-engine materials including superalloys,titanium-based alloys,composites and thermal barrier coating.Considering the extreme service environment of aero-engine materials,the review is concluded by discussing the possible challenges and solutions.

关 键 词：材料基因工程 机器学习 主动学习 航空发动机材料 高温合金 

分 类 号：V26[航空宇航科学与技术—航空宇航制造工程] V25[一般工业技术—材料科学与工程]