Research On Construction Risk Assessment Of Xiangxi Yangtze River Bridge B Ased On Dynamic Multi-source Monitoring Data

Cable hoisting-cable shackle is an important construction method for steel arch bridges,but this construction method will bring complex internal force and displacement changes to the bridge structure.The risk during the construction process has been gradually exposed,in order to ensure the quality and safety of bridge construction,it is urgent to analyze its construction risk.Based on the Bayesian network theory and the dynamic multi-source monitoring data,the risk of steel arch bridge cable hoisting construction is analyzed in this paper.Firstly,according to the construction characteristics of cable hoisting-cable shackle and the main contents of steel arch bridge construction monitoring,the influencing factors of steel arch bridge construction risk are analyzed,and the elastic modulus change,anchor cable tension error,anchor tower offset,anchor cable measurement error and temperature change are proposed.Taking the Xiangxi Yangtze River Bridge as an example,the finite element model of the arch rib cable hoisting is established,and the influencing factors are analyzed.The calculation results show that the anchor cable tension error and measurement error have the greatest influence on the vertical deformation of the arch rib,and the maximum sensitivity coefficient of the arch rib shape is 450.Therefore,the tension of the anchor cable has a great influence on the construction risk of the steel arch bridge.It needs to be strictly controlled in the process.Secondly,based on the application of Bayesian network in analyzing the uncertainty problem,the Bayesian network topology structure of the cable hoisting construction risk of Xiangxi Yangtze River Bridge is constructed with the arch rib closure error as the target value.According to the dynamic multi-source monitoring data of Xiangxi Yangtze River Bridge,the mean and variance of the normal distribution of the root nodes of the topology are determined.According to the probability distribution of the root node and the finite element model of the ribbed cable hoisting,the BP neural network and Monte Carlo method are used to obtain the probability distribution of the arch rib closure error.Then the Bayesian network calculates the probability of the error of the arch ribs of the Xiangxi Yangtze River Bridge,and combines the "Guidelines for Safety Assessment of Highway Bridge and Tunnel Engineering Design" to evaluate the error risk of the arch ribs.The results show that the probability of the arch rib closure error is greater than 60 mm reaches 0.2327,at the same time,the error of the arch ribs is up to the IV level at the risk level of 60~120mm.It is necessary to monitor the tension of the anchor cable,the deformation of the tower and the temperature change in real time.Finally,the vertical deviation probability of the arch rib structure of the Xiangxi Yangtze River Bridge is calculated according to the construction stage.The results show that the risk probability is transitive,and the risk probability of the deviation value between 0 and 40 mm will be transmitted to the probability of more than 40 mm with the progress of the construction.The probability that the vertical deviation of the arch ribs is greater than 40 mm increased from 0.0903 to 0.3507.Based on dynamic multi-source monitoring data and Bayesian network,this paper establishes a steel arch bridge cable hoisting construction risk assessment model,and verified it through engineering practice.It is expected to provide useful reference for analyzing and evaluating similar bridge construction risks.It has certain theoretical reference and engineering application value.