We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Analysis and prediction of shrinkage cavity defects of a large stepped shaft in open-die composite extrusion based on machine learning.
- Authors
Wang, Menghan; Du, Menglong; Li, Songlin; Wang, ZhouTian
- Abstract
A stepped shaft, as an integral part of an aeroengine system, is prone to shrinkage cavity defects during open-die composite extrusion, which affects the service performance and life of the fan shaft. First, the deformation process of the fan shaft in open-die composite extrusion was analyzed using finite element simulation. The shrinkage primarily occurred in the forward extrusion stage. Subsequently, the mathematical conditions of the shrinkage cavity in the forward extrusion were obtained using the differential element method. The die parameters affecting the shrinkage cavity were mainly the die inclination and extrusion ratio. A finite element simulation of a simplified forward extrusion model and a machine learning classification algorithm were used to create a shrinkage cavity prediction diagram with solid performance and good generalization. Stepped-shaft forging with satisfactory performance and no shrinkage was obtained by selecting appropriate die parameters according to the shrinkage prediction diagram.
- Subjects
MACHINE learning; FRUIT drying; CLASSIFICATION algorithms; FORECASTING
- Publication
International Journal of Advanced Manufacturing Technology, 2023, Vol 127, Issue 5/6, p2723
- ISSN
0268-3768
- Publication type
Article
- DOI
10.1007/s00170-023-11634-4