Progressive assessment of water inrush disasters in pre-construction and construction phases based on Chinese tunnels research

The frequency of water inrush disasters severely affects tunnel construction, lives, and property. As a result, accurate forecasting of the danger of water inrush during construction is critical. This paper aims to create a progressive evaluation model for assessing water inrush risk during two stages: pre-construction and construction. The proposed model provides a static pre-construction estimation and allows dynamic updates based on previous predictions for unexcavated sections during construction. Initially, comprehensive multi-water inrush information was presented by analyzing 65 tunnel accidents of water during construction, including 12 evaluation indexes encompassing hydrogeology and excavation monitoring. Subsequently, the assessment model was constructed using a combined weighting method and non-linear attribute recognition theory. The information from dynamic monitoring on surrounding rock and seepage pressure is integrated, and feedback site data is used to dynamically modify the weighting of the indicators and update the projected predictions. This approach has been employed as a case study to evaluate the water inrush risk during the Qinling water transmission tunnel. The results demonstrate that the prediction outcomes of dynamic weights align well with on-site holes, exhibiting higher forecasting accuracy than other methods. This approach offers a novel perspective for accurately evaluating water inrush risk in tunnel construction.