Research On Temperature Effect Separation And Deformation Prediction Of For Long-span Cable-stayed Bridges Based On Monitoring Data

Bridge structures are affected by various external loads and actions during operation.For long-span steel cable-stayed bridges,because they belong to multiple statically indeterminate structural systems and steel has good thermal conductivity,the structural deformation and internal force responses under temperature can not be ignored,sometimes even exceed the structural response caused by vehicle loads.When evaluating and warning the technical status of long-span steel cable-stayed bridges,in order to avoid the real response of structures under vehicle loads being "masked" by temperature response,the temperature effect in various structural responses should be reasonably eliminated.The structural response caused by temperature makes the change of the total response more complex,so it is important to master the rule of structural response changing with temperature.Based on the measured data in the long-term monitoring system of Nanjing Yangtze River Bridge,this paper takes the temperature effect of the main girder deflection as the main research object,counts the deflection and temperature effect of the main girder,analyses its variation law,and inspects the influence of temperature effect on the main girder deflection.Secondly,based on ABAQUS software,the finite element model of steel box girder segment is established,and the temperature effect of steel box girder under sunshine temperature is simulated and analyzed.Based on MIDAS/Civil software,the full bridge model of the background bridge is established,and the temperature sensitivity of the main girder deflection under various temperatures is analyzed.Then,based on the multi-scale analysis method,combined with the time and multi-scale characteristics of the deflection of the main beam,the temperature effect of the deflection of the main beam is effectively separated in annual,seasonal and daily cycles,and the quantitative fitting between the daily temperature effect of the deflection of the main beam and the atmospheric temperature in sunny weather is realized by using the multi-fold line model.Finally,combined with the finite element analysis results of the sensitivity of temperature to the deflection of the main girder,the artificial neural network is constructed to fit and predict the daily temperature effect of the main girder deflection with the input of time,atmospheric temperature and the vertical temperature gradient of the steel box girder and the output of the daily temperature effect of the main girder deflection.