Dynamic prediction for tunnel water inflow based on non-Darcy flow theory with advanced geological prediction information fusion

Accurate prediction of water inflow is critically important in engineering for preventing and mitigating water inrush disasters during tunnel construction. This study introduces a novel dynamic prediction model that integrates multi-source advanced geological forecast data and proposes a water inflow prediction methodology grounded in non-Darcy flow theory. The developed model effectively captures the nonlinear characteristics inherent in water inrush processes and accommodates the spatiotemporal evolution of hydrogeological conditions throughout tunnel excavation. Key parameters for water inflow computation were derived from advanced geological prediction data, while dynamic forecasting was accomplished through the combined application of groundwater dynamics theory and entropy-weighted analysis. The results demonstrated that static water inflow exhibited an exponential relationship with seismic wave velocities, the permeability coefficient followed a power-law function with apparent resistivity, and dynamic water inflow decayed following a negative exponential function. The comprehensive water inflow showed positive correlations with both the permeability coefficient and groundwater depth while exhibiting negative correlations with the non-Darcy flow coefficient and decay rate. Furthermore, the decay rate of water inflow was positively correlated with the decay coefficient. A comparative analysis with existing methods, field monitoring data, and numerical simulations confirms the effectiveness of the proposed method. This method has been successfully implemented in a mountain tunnel project in Southwest China, with prediction errors maintained below 5%. The proposed method provides valuable technical support for the early warning and risk management of water inrush hazards in tunnel engineering.