Safety Risk Analysis Of Subway Construction Based On Fault Tree Method

With the rapid development of China's economy, subways are being or will be constructed in most large and medium cities. Due to the long construction period, dynamic design scheme and difficult management, the risk of subway tunnel construction has the characteristics of diversity and multilevel, In addition, the scientific evaluation tool and management system of risk have not been formed, the contingencies usually occur in subway tunnel construction. In this dissertation, Fuzzy Fault Tree mathod and Bayesian Network Fault Tree method are employed to systematically study the risk of construction cost of shield tunnel, failure risk of segment of shield tunnel and failure risk of deep foundation pit of subway station.First of all, the fundamental principles and analysis procedure of fault tree method, as well as the general principles and specific steps of establishing a fault tree are systematically introduced. The logic gates and system structure function expression of the fault tree are also explained. The qualitative analysis method of fault tree describes uplink and downlink methods for solving the minimum cut set of the fault tree, the quantitative analysis method of fault tree explain the basic principles and calculation methods of importantance degree of events. Fuzzy theory is introduced to eliminate the limitations of the traditional fault tree. The theory of fuzzy sets and fuzzy membership, fuzzy set operations, expansion theory, convex fuzzy set andλcut set, L-R type fuzzy number and other relevant knowledge are also introduced, and fuzzy operator of fault tree analysis are given as well. Fuzzy numbers are used to describe the probability of the basic events in the fault tree, the problem that the precise probabilities of the basic events cannot be obtained is solved. The fuzzy gate operator used instead of the logic gate operator results in the fuzzy numbers of probability of occurrence of the top event, which can reflect the actual situation better. Finally, two kinds of commonly used fuzzy importance degree and their calculation methods are introduced.Secondly, the fault tree method based on triangular fuzzy numbers is applied to analyze the risk of construction costs of shield tunnel of subway, and the corresponding cost importance degrees are calculated and obtained. The calculated results obtained from the fault tree method based on triangular fuzzy number, the traditional fault tree method and economic loss multiplication method are compared, and the results show that:①the impact of various risk factors on the whole project, the fuzzy characters of bottom event probability are both considered in triangular fuzzy numbers method, and the results are relatively accurate.①during the construction period of a shield tunnel, the two most severe risks to total cost are "leakage of water and grout when the shield enters and leaves" and "encountering obstacles during advancing", which should be paid attention to and evaded by carrying out effective measures.Thirdly, the fuzzy fault tree method is employed to investigate the risks of the shield tunnel segment floating, the dislocation and cracking failure in subway shield tunnel, and the calculation results show that:①based on fuzzy number representation theorem, the exact solving method for multiplication of fuzzy numbers is derived, it is a solid foundation for the application of fuzzy theory in fault tree method.①according to the requirement of the important degree of the structures, the "shield machine advancing posture" and "grouting slurry parameters" risks should be avoided at first. The other three risks "poor geologic condition of water-bearing strata in hard rock", "cracking of segment extrados" and the "uneven jacking force between segments", although their probability impacts are small, the importance degrees of structures are great. Thus commensurate construction standards and appropriate protective measures should be adopted.③when the exact solving method based on the representation theorem is employed to calculate the multiplication of the triangular fuzzy number, it can make the results more accurate. Meanwhile, the confidence index and the triangular fuzzy number are used to modify the experts rating, and the obtained results have high accuracy and the method is simple to use. So this method is highly recommended.Finally, fault tree analysis method based on Bayesian network is employed to analyze the construction risk of deep excavation of the subway station, the results show that by using Bayesian network model on the deep excavation system, more accurate results are obtained and the complexity of the expression and the calculation speed are significantly superior to the traditional fault tree model, and it is easy to find the weaknesses in the system. Case study on deep fondation pit of Xunlimen station of Wuhan metro line 2 shows that the fault tree analysis method based on Bayesian network is more convenient and practical than the ordinary fault tree analysis method. Meanwhile, the problems in calculation scale and post adjustment of the ordinary fault tree model are also solved.The achievements In this dissertation can be valuable references for risk analysis and control of shiel tunnel in subway construction.