| At present,some research results have been obtained for the risk assessment of the construction stage of rigid frame bridge,including the analysis and research of the construction state.However,at present,the structural safety risk analysis methods for the cantilever construction stage of rigid frame bridge are mostly based on qualitative considerations and subjective.In addition,the existing risk assessment theory can not consider the randomness of risk factors and the existing construction monitoring data are not well used in the risk assessment.These problems lead to the low reference value of the current bridge risk assessment results.Based on the above background,this paper studies the structural safety risk assessment method of cantilever construction stage of rigid frame bridge based on statistical inversion theory.The main contents are as follows:(1)Based on the theory of reliability and risk assessment,the structural safety risk of rigid frame bridge in construction stage is classified.First,the five stage model of risk assessment is described.Then,based on the reliability theory,combined with the limit value of structural index in relevant literature,and the structural safety risk in cantilever construction stage of rigid frame bridge is divided into five levels.(2)The structural safety risk of rigid frame bridge in cantilever construction stage is identified.Based on the engineering example of Tongzi River Bridge,this paper firstly establishes the finite element model of Tongzi River Bridge and calculates the structural response of No.3 pier under the maximum cantilever state;secondly,combined with the characteristics of the cantilever construction stage of Tongzi River Bridge,the elastic modulus of concrete,shrinkage and creep,self weight of structure and prestress of steel tendons are selected as risk analysis objects;finally,the risk factors are obtained The sensitivity coefficient of element to structural response.The results show that: the structural response is the most sensitive to the structural self weight,followed by the tendon prestress,concrete elastic modulus,concrete shrinkage and creep.(3)The PSO-BP neural network model is used as the proxy model of the finite element model.To reduce the massive computational cost of structural security risk analysis,this paper constructs a neural network model based on PSO-BP algorithm to predict the structural response under different combinations of risk parameters,and implements the algorithm using Python3.7 programming.The results show that the model has higher prediction accuracy and faster convergence than the BP neural network model,and it is feasible as a proxy computing model of finite element model.(4)Based on the statistical inversion theory and risk classification standard,the structural safety risk of Tongzi River Bridge in the cantilever construction stage is analyzed.In this paper,the construction stage of the No.7 block of the middle span side tension of the No.4 pier of Tongzi River Bridge is selected.Taking the vertical deformation deviation of the main beam as the index,the risk probability of structural safety in the cantilever construction stage is calculated and compared by Gibbs sampling and Monte Carlo method respectively under the prior and posterior distribution of key risk factors,and the risk classification is carried out in combination with the risk classification standard.The results show that the posterior probability distribution of the risk factors of Tongzi River Bridge is more in line with the actual situation of the project,which can provide a reference for the risk management of the cantilever construction stage of Tongzi River Bridge. |