基于图神经网络的在役隧道爆炸响应快速分析方法及系统
By using a graph neural network-based approach, tunnel degradation patterns are quantified as graph structure attributes. A graph neural network surrogate model is constructed to quickly generate a scenario-response database, solving the problem of inaccurate prediction of explosion responses in in-service tunnels. This achieves efficient and accurate analysis results, providing a reliable basis for tunnel explosion-proof protection design.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- JIANGHAN UNIVERSITY
- Filing Date
- 2026-05-14
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies are insufficient to accurately predict the explosion response of in-service tunnels caused by material performance degradation, segment joint deterioration, and interface defects. Furthermore, traditional methods are time-consuming, labor-intensive, lack physical interpretation, and have limited generalization capabilities.
By using a graph neural network-based approach, tunnel degradation patterns are quantified as graph structure attributes. A graph neural network surrogate model is constructed, and combined with efficient sampling and parallel computing, a scenario-response database is quickly generated. Damage features are then reduced in dimensionality and clustered to reveal damage patterns and evolution paths.
It enables efficient and accurate prediction of the explosion response of in-service tunnels, improves the physical consistency and accuracy of the analysis, and provides a reliable basis for explosion-proof protection design.
Smart Images
Figure CN122197167B_ABST