Risk Assessment And Management Of Leakage In Operation Shield Tunnel

With the economic development and population growth,China's rail transit has witnessed rapid development in recent years.Due to the special structural of the shield tunnel and the complex occurrence environment,there is a high chance of diseases development in operation shield tunnels.As one of the most common diseases in operation shield tunnels,seepage can not only aggravate other diseases,but also also threatens the safe operation of tunnels caused by subway train outages and subway circuit damage.Therefore,the safe operation of shield tunnel should be guaranteed through the risk assessment and management of seepage during the operation period of shield tunnel.Based on the risk mechanism analysis of tunnel waterl eakage,a three-level index system for leakage risk assessment is constructed,which includes 4 secondary indexes and 11 tertiary indexes.For the leakage risk system with complexity and nonlinearity,a PCBN model based on Bayesian and Copula theory is proposed to evaluate the spatial dependence of operational tunnel leakage risk.And the PCBN model is applied to Wangzong section tunnel of Wuhan Metro Line 3.The third-level indicators' data is collected from multiple monitoring points with different spatial distributions of the tunnel.Then the optimal marginal distribution and correlation coefficient representing the spatial distribution of third-level indicators' data are determined.The PCBN model for this tunnel is constructed and verified to calculate and determine that the leakage risk status of this tunnel is basic safe.The spatial correlation analysis is carried out on the three-level indexes to determine that the Seal breakage and aging rate and Bolt failure efficiency are the indexes with greater influence on the leakage of the tunnel,which are used as key risk factors for decisionmaking to reduce the risk of water leakage.For Wang Zong tunnel's one empty push segment with high water leakage risk,small sample data and multi-source information,a method based on SVM-DS theory is proposed to evaluate the water leakage risk.Based on the monitoring data sets,the SVM is used to obtain the secondary indexes' BPA in test set and DS theory is used to determine the test set of the individual monitoring-station seepage risk status.Measures to reduce the risk of water leakage are formulated for the monitoring points with higher risk of water leakage in the evaluation results.On this basis,it is used as the target monitoring point to strengthen monitoring,to obtain the three-level index time series monitoring value,to explore the effectiveness of risk control measures,and to put forward the prediction method of time dependence of leakage risk based on SVM,cloud model and DS theory.Based on the same sample data,using the SVM time correlation analysis,get the time evolution law of three-level indicators measured values,thus get threelevel index forecast time series prediction,based on the cloud model is adopted to solve forecast fuzzy uncertainty,obtaining the DS after BPA level 3 indexes,risk level and the trend to predict target monitoring the leakage,to determine in seepage control measures under the action of target monitoring risk evolution has lower over time,to demonstrate the effectiveness of control measures,provide a reference for the tunnel seepage risk management.