.png){kind=icon}
.png){kind=icon}
Return
Implicit-PointSAGE: Neural Implicit Representation based Superresolution for computational fluid dynamics
.png){kind=icon}
PRE
AI
DOI：10.1016/j.compfluid.2025.106900.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Learns fine flow features from sparse simulations using point-based deep learning. • Introduces physics-aware attention for super-resolution of fluid dynamics. Combines local and global flow patterns through novel point cloud modeling. • Achieves mesh-free high-resolution predictions without structured grids. • Demonstrates strong accuracy–efficiency trade-off on complex flow cases.
Journal
C
IF:
3
Papers:
333
Citations:
1.4W