AI-Assisted Numerical Analysis of Gasket Deformation and Contact Stress in Shield Tunnel Waterproofing Systems

Abstract

Waterproofing in shield tunnels is critical for ensuring structural integrity and long-term durability, especially in large-diameter tunnels where joint and segment waterproofing play essential roles. This study employs a combination of numerical simulations and AI-assisted analysis to evaluate the deformation characteristics and contact stress distribution across six gasket cross-section designs. Through finite element analysis (FEA) and machine learning-driven pattern recognition, this research identifies key deformation behaviors and stress concentration areas. Results demonstrate that, under identical hardness conditions, the contact stress at the gasket-gasket interface has the most significant impact on waterproofing performance, surpassing the gasket-pad and gasket-groove interfaces. Section A exhibits optimal waterproofing capability and stable contact stress growth, making it the preferred design for engineering applications. The incorporation of AI techniques further enhances the prediction accuracy and design optimization process, providing a data-driven framework for improving gasket designs in shield tunnel systems.