|
|
|
| Progress of Data-Driven Aero-Engine Materials Design |
| 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) |
|
|
|
|
Abstract 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.
|
|
|
|
|
|
|
|
|
2021, 
