Physics-informed edge-enhanced temporal graph convolutional network for multi-risk evolution prediction in deep excavation

Key Points

• The model significantly enhances risk evolution prediction, addressing multiple risks involved in deep excavation.
• An 80% accuracy rate is achieved in predicting risks over various time points during excavations.
• Assessment using a physics-informed edge-enhanced temporal graph convolutional network offers innovative insights into risk management.
• Improved prediction methods may enable better planning in excavation projects, crucial for safety and efficiency.