地铁深基坑施工耦合风险动态管控研究

李晨爽

摘 要

目前，我国地铁建设发展迅速，与此同时，地铁工程安全事故也不容忽视。地铁深基坑施工受地质条件，周边环境等众多因素影响。一方面，安全风险因素之间存在相互影响关系，另一方面，随着施工进度的推进 ，这些相关因素也处于不断的动态变化中，因而地铁深基坑工程容易发生风险事故。因此，本文针对深基坑风险因素耦合关系、风险传递及动态分析方法相关问题展开研究，提出深基坑耦合风险动态分析模型，并应用于某市地铁车站基坑工程，从结果来看，该模型能够有效地对耦合风险进行动态分析，实现动态风险管控。首先，针对深基坑耦合风险动态管控现状，本文从两个方面对国内外理论进行重点分析，即地铁深基坑风险管理和动态风险耦合模型。其次，基于以上理论，本文构建了深基坑耦合风险动态管控模型：（1）针对风险因素之间的关系，利用Apriori算法进行关联规则挖掘，进行风险因素关联分析；（2）在以上基础上进行基于复杂网络的耦合风险分析，发掘基坑耦合风险演化规律；（3）基于关联规则和复杂网络的结果，构建基于贝叶斯网络的深基坑耦合风险动态管控模型，对深基坑风险进行施工前风险预测，施工中关键因素控制以及事后事故致因诊断，并提出风险管控措施相关建议，从而实现基坑施工耦合风险的动态管控。最后，以某市地铁车站为例，将上面的模型运用到具体地铁施工中，验证其有效性和适用性，最终结果表明该模型可行并具有应用潜力，为地铁车站基坑施工过程中的安全管控提供了有效的决策支持。

关键词：地铁深基坑；耦合风险；动态管控；关联规则挖掘；复杂网络；贝叶斯网络

Abstract

At present, China's subway construction has entered a golden period of development. With the rapid development of subway projects, subway safety accidents have also occurred frequently. The construction of subway foundation pits is affected by many factors such as geological conditions and surrounding environment. On the one hand, risk factors are not independent and they are interrelated. On the other hand, these interrelated risk factors are constantly changing during the construction. Therefore, there are many risks during the deep foundation pits construction of subway stations. Therefore, this paper studies the coupling relationship, risk transfer law and dynamic analysis method of subway engineering risk factors, and proposes a dynamic analysis model of deep foundation pit coupling risk. In addition, the model is also applied in the foundation pit engineering of a metro station. The engineering verification shows that the method can effectively deal with the complex coupling relationship and achieve dynamically risk control.

Firstly, in order to achieve dynamic control of coupled risk in deep foundation fit, this paper focuses on the analysis of relavant reseach from two aspects: the risk management of deep foundation pit and dynamic risk coupling model. Secondly, based on the above theory, this paper constructs a dynamic control model for deep foundation pit coupling risk:(1) the risk association analysis is carried out based on Apriori algorithm of association rules minning, and the correlation between risk factors is explored;(2)the coupling risk analysis based on complex network is carried out to find the law of foundation pit coupling risk transfer;(3)based on the association rules and the results of complex networks above, the dynamic Bayesian network model of deep foundation pit in coupling risk control is constructed. This model can achieve real-time risk analysis, including pre-accident, during-construction and post-accident control. And relevant recommendations for risk management and control measures are put forward to achieve dynamic coupling risk control to ensure the construction safety of deep foundation pit. Finally, the above model is applied to a deep fit foundation in a subway station to verify its effectiveness and applicability. The result shows that the model is feasible and has potential application. Safety control during the construction of the foundation pit provides effective decision support.

Key words: Deep metro foundation pit; Coupling risk; Dynamic risk control; Association rules mining; Complex network; Bayesian network