基于动态贝叶斯网络的地铁隧道施工风险评估

Risk assessment of metro construction based on dynamic Bayesian network

近年来中国地铁隧道建设发展迅速,然而地铁施工事故频繁发生。针对施工过程中风险随时间的演变,提出基于动态贝叶斯网络(DBN)和模糊综合评价法(FCEM)的地铁隧道施工动态风险评估方法,结合沉降监测数据进行地铁隧道下穿既有建筑物的风险实时动态评估。该方法通过施工环境、施工技术、施工机械、施工管理4方面共12个风险因素与施工风险间的因果关系,建立地铁隧道施工风险DBN模型,预测施工过程风险发生概率的动态变化;根据施工实测的累计沉降值和沉降变形速率反馈DBN模型,更新动态风险概率评估结果;基于FCEM评估地铁隧道施工风险损失,并计算施工动态风险值。以北京地铁14号线地铁盾构施工下穿工程为案例验证方法的合理性,结果表明该方法结合监测数据的动态风险评估结果能较好地反映实际施工过程中的突发风险事件。

Recently,tunnel construction has developed rapidly in China.However,tunnel construction accidents occur frequently.In order to reveal the risk evolution process during the construction process over time,a dynamic tunnel construction risk assessment method is proposed based on the dynamic Bayesian network(DBN)and the fuzzy comprehensive evaluation method(FCEM).The settlement monitoring data induced by underpass tunnel excavation can be used to evaluate the real-time risk of above-ground structures by employing the proposed method.Twelve risk factors are classified into construction environment,construction technology,construction machinery and construction management.The causal relationship between twelve risk factors and tunnel construction risk is considered to build the DBN model,and the model is used to predict the evolution of risk during the construction process.The monitoring data of settlement and subsidence rate are used to update the assessment results of dynamic risk probability.The tunnel construction risk is evaluated by the fuzzy comprehensive evaluation method and the construction dynamic risk is calculated by the definition of risk.The reasonability of the proposed method is verified through the case of shield construction project of Beijing Metro Line 14.The results show that the proposed method can reflect the unexpected risk events in the actual construction process.