A novel cloud model for risk analysis of water inrush in karst tunnels

Abstract

Water inrush is a serious geological hazard in underground engineering. The prediction of possibility and classification of water inrush risk has long been a global problem for the construction of deep-buried tunnels in karst areas. To solve the randomness and fuzziness in the evaluation process of water inrush risk, a novel comprehensive evaluation model was established based on the normal cloud theory. According to the systematic analysis of the influence factors of water inrush, seven factors were selected as evaluation indices, including formation lithology, unfavourable geological conditions, groundwater level, landform and physiognomy, modified strata inclination, contact zones of dissolvable and insoluble rock, and layer and interlayer fissures. Meanwhile, a hierarchy model of the influence factors was established for water inrush, and the analytic hierarchy process was adopted to determine the weighting coefficients for each evaluation index. The normal cloud theory was used to describe the cloud numerical characteristics for each evaluation index of risk classification for water inrush. Normal cloud droplets were generated to reflect the uncertain transformation between the risk levels of water inrush and the evaluation indices. Then, the synthetic degrees of certainty were calculated, and risk level of water inrush was determined. Finally, the proposed model was applied to two typical deep-buried tunnels in karst areas: Jigongling tunnel and Xiakou tunnel. The obtained results were compared with the relevant analysis results and the practical findings, and reasonable agreements were gained. The normal cloud model was found to be more accurate, feasible and effective for risk classification of water inrush prediction. It can not only meet the requirement of tunnel engineering, but also be extended to various applications.