| In the harsh environmental and complex loading conditions over a long-term service period,the key structural components of long-span bridges inevitably sustain fatigue and corrosioninduced damage due to long-term vehicle load,wind load and other dynamic loads,as well as environmental erosion.As a result,the structural performance of full-bridge deteriorates,even endangers the service safety of the bridge.As the typical diseases of the key structural components of long-span bridges,fatigue and corrosion condition assessment and life prediction for the existing long-span bridges has always been a hot research issue in the field of structural condition assessment.The fatigue damage is often the result of the coupling effect of various loads and factors,which initiates from the local details of the structures,and then gradually propagates,it has obvious size effect.For the welded structure,welding residual stress also has great influence on the fatigue performance of welded structure.As one of the most common and important load-bearing components with the shortest service life in long-span bridges,the fatigue resistance and plasticity of stay cables will be significantly reduced,and the brittleness will be enhanced under the coupling action of vehicle load,wind load and corrosive environment.As one of the dominant factors of fatigue life of steel bridges,vehicle load is the load basis of safety degradation and residual life prediction of in-service bridges.Therefore,the research objective of this paper is to propose a fatigue assessment and life prediction method for steel bridges considering the coupling effect of in-situ traffic loads,ambient temperature,welding residual stress and corrosion.Based on the long-term monitoring information monitored by the bridge health monitoring system,the in-depth study has been carried out from four aspects:in-situ traffic loads,environmental temperature effect,multi-scale finite element modeling and model updating method,and fatigue condition assessment method considering multi-factor coupling effects.The main contents and contributions of this dissertation are listed as follows:(1)Vehicular fatigue load spectrum and stochastic process model of traffic flow for highway bridge.Firstly,the statistical analysis of the traffic flow loads across the bridge is carried out based on the traffic flow monitoring information monitored by the weigh-in-motion system,and the corresponding vehicle load spectrum and the random process model of in-situ traffic flow loads are established.Under the condition that the traffic flow,composition of traffic flow,arrival time and individual characteristic information of related vehicles at any time are known,the cellular automata method is used to simulate the evolution process of the traffic flow on the bridge,in order to obtain the specific distribution of vehicles on the bridge at any time,and then to provide a load analysis model for the fatigue condition assessment and life prediction of the bridge.(2)Probability model of ambient temperature and its influence on fatigue performance of steel decks.The vehicle load dispersal effect and the composite stiffness effect between asphalt concrete pavement and bridge deck caused by asphalt concrete pavement under different temperature conditions have been studied for the further study of the influence of temperature on the fatigue performance of steel bridge deck.Based on the monitoring information of environmental temperature at the bridge site,the bridge deck temperature and the structure temperature,the probability distribution model of the environmental temperature,the relationship model between the bridge deck pavement temperature and the environmental temperature,the relationship model between the structure temperature and the environmental temperature are established respectively.(3)Bridge multi scale finite element modeling and monitoring data based twin model updating method.The key to the fatigue condition assessment is the establishment of multi-scale finite element model considering structural damage size effect.However,there must be some differences between the actual structure and the finite element model due to the influence of the size error,material property error,boundary condition error,model analysis algorithm error and model order error.The existence of this difference will seriously affect the accuracy of the fatigue condition assessment and the life prediction results.Therefore,the corresponding multi-scale finite element model updating methods are proposed(response surface-based finite element model updating,multi task joint optimization network based multi scale finite element model updating and multi-modal fusion regression network based multi scale finite element model updating).The single objective optimization problem in the traditional model updating is transformed into the Pareto optimization problem corresponding to the super multi-objective function in the prosed methods for the multi scale finite element updating so as to ensure the high fidelity of finite element model.(4)Fatigue assessment of orthotropic steel deck in cable-stayed bridge considering traffic flow and temperature actions.Using the data monitored by the structure health monitoring system of a Yangtze River Bridge,including the structural strain response information,traffic flow information and temperature information of bridge deck pavement,a fatigue damage regression model is established,which can consider both vehicle load and ambient temperature.And then,a fatigue reliability evaluation method based on the fatigue damage regression mathematical model is further proposed.Finally,it is applied to fatigue life prediction of orthotropic steel deck structure of the cable-stayed bridge.Different from the traditional evaluation method based on stress-life relationship,this method can not only consider the influence of the coupling effect between vehicle load and ambient temperature on the fatigue performance of steel deck,it can also quantify the effects of loading order and traffic flow growth mode on fatigue life.(5)Digital twin-based health hybrid monitoring and fatigue evaluation of orthotropic steel deck in cable-stayed bridge.Welding residual stress is an important cause of fatigue cracking for orthotropic bridge deck.Under the repeated traffic load,cracks will gradually initiate and expand from the tensile region of welding residual stress.It is difficult to accurately assess the fatigue state of orthotropic steel deck only based on the existing traffic load monitoring information,while ignoring the coupling effect of traffic load,ambient temperature and welding residual stress.Therefore,the sub-model technology is employed to study the distribution characteristics of welding residual stress and the coupling effect with vehicle induced stress and temperature induced stress in the welding seam between the upper deck and U rib of orthotropic bridge deck.Aiming at the defect that can’t consider welding residual stress for the S-N curves based fatigue condition assessment and life prediction method,a nonlinear fatigue damage model based on continuum damage mechanics,which has been verified by orthotropic bridge deck fatigue test,is used to evaluate the welding seam between the upper deck and the U-rib of the orthotropic steel desk for an in-service cable-stayed bridge.(6)Fatigue assessment of stay cables in cable-stayed bridge considering in-situ traffic flow loads and corrosive environments.The corrosion of high-strength steel wire in the cable body of in-service parallel-wire stay cables will aggravate,the fatigue resistance and plasticity of stay cables will be significantly reduced,and the brittleness will be enhanced under the coupling action of vehicle load,wind load and corrosive environment.Then the stay cables would suffer from serious corrosion and fatigue disease and lose their service function prematurely.The fatigue life distribution model of high strength steel wire considering the effect of pretension is firstly established using the multi-parameter Weibull model based on the fatigue test results of grade 1670 high strength steel wires with different corrosion states.Then,based on the fatigue life distribution model of single high-strength steel wire and the parallel system theory,the fatigue life model of stay cable considering the coupling effect between cable tension force and corrosion is established by Monte Carlo simulation method.Finally,combining the established random vehicle load model,the fatigue state of stay cables under different corrosion states is evaluated. |
PDF Full Text Request