.png){kind=icon}
.png){kind=icon}
Return
Hierarchical graph representation and interpretable operational carbon emission assessment of residential spatial layouts using Graph Residual Fusion Network: An automated design support framework
.png){kind=icon}
PRE
AI
J
L
W
H
X
Z
K
DOI：10.1016/j.aei.2026.104723.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A hierarchical graph representation method for architectural spatial layouts. • A novel GRFN architecture for accurate building carbon emission prediction. • A large-scale OCE-labeled high-rise residential graph dataset across three climate zones in China. • GRFN models outperform MLP and XGBoost in accuracy, generalization, and robustness. • Interpretable GNN-driven design support framework facilitates low-carbon decision-making.
Keywords:
Hierarchical graph representation
Graph Residual Fusion Network
Operational carbon emission assessment
Interpretable GNN
Residential spatial layouts
Journal
IF:
9.9
Papers:
4.0K
Citations:
1.7W
Organization
Cited Papers