This study aims to develop a high-accuracy life prediction model for Tunnel Boring Machine (TBM) disc cutters that overcomes the limitations of existing models, specifically their inability to quantitatively couple multiple wear mechanisms and insufficient refinement of kinematic parameters.
A dynamic wear prediction model integrating relative slip distance and multi-mechanism synergy is proposed. The model quantifies the contributions of abrasive, adhesive and fatigue wear through weighting coefficients (a, b, c). The relative sliding distance (L_s) is dynamically calculated based on cutter kinematics. The model was validated using field data from a TBM project in mixed granite strata, and its sensitivity to the abrasive wear coefficient (K1) was analyzed.
The model demonstrated high accuracy, with maximum prediction errors of 12.2% and 13.9% in two tunneling sections, and a coefficient of determination (R²) of 0.78–0.86. Abrasive wear was identified as the dominant mechanism, accounting for 94.9% of the total wear. Sensitivity analysis confirmed K1 = 4 × 10−3 as the optimal coefficient. The model outperforms traditional models in dynamic parameter adaptability and offers greater interpretability than pure data-driven models.
This research presents a novel dynamic weighting framework that integrates multiple wear mechanisms with a refined kinematic analysis of relative sliding distance. This approach addresses a critical gap in existing models, moving beyond single-mechanism assumptions and coarse parameterization. The model achieves high predictive accuracy without relying on extensive training data, providing a robust and interpretable tool for intelligent cutter management in TBM projects.
