A New Bayesian Network Model for the Risk Assessment of Water Inrush in Karst Tunnels

Water inrush seriously restricts the safe construction of a karst tunnel. Once it occurs, will cause serious consequences such as economic loss and casualties. Due to complexity an underground environment, is difficult calculate probability tunnel water inrush. Therefore, great engineering significance establish effective risk assessment model. Based on Bayesian theory, interpretation structure model, generative adversarial network, network model established. The results show that firstly, twelve indexes selected can not only characterize but also are easy be counted, which effectively improves accuracy Secondly, overcomes shortcomings other models rely too much geological data through massive training. Thirdly, corresponding noninrush samples generated by analytic hierarchy process, solve problem unbalanced database. Finally, applied DK490+373 section Shangshan Tunnel. operable, effective, practical, suitable for situation incomplete index statistics.