.png){kind=icon}
.png){kind=icon}
Return
Parallel kernel deep learning model for damage quantification in laminated composites
.png){kind=icon}
PRE
AI
M
M
H
DOI：10.1016/j.advengsoft.2026.104168.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A Parallel Kernel CNN extracts both global and local Lamb-wave features simultaneously, achieving 95.83% severity assessment and 98.15% localization accuracy in laminated composites. • The model eliminates kernel-specific tuning and provides a unified framework for damage severity assessment and localization. • Experimental validation demonstrates superior performance across varying damage conditions.
Keywords:
Parallel Kernel CNN
Lamb-wave features
damage severity assessment
localization accuracy
laminated composites
Journal
IF:
5.7
Papers:
3.3K
Citations:
1.2W
Organization
No organization information available
Cited Papers