Time-dependent Reliability Analysis And Damage Identification Method For Existing Medium And Small Span Bridges In Seasonal Frozen Region

As one of the important infrastructures,bridge structure plays an important role in rapid social development,of which the safe operation is associated with national economic development healthily and security of people’s life and properties.Medium and small span bridges occupy a high proportion in the existing bridge system due to their easy installation and wide application.However,because of influence of unfavorable environmental factors and increasing vehicles load,bridge performance is degraded and different degrees of damages appear.It brings great hidden dangers to normal traffic operation and personal safety.Therefore,evaluating the performance of existing bridges accurately can not only provide a strong basis for daily maintenance and reinforcement,but also avoid bridge safety accidents effectively.Reliability theory has been widely used for bridge performance evaluation,which can quantitatively evaluate the ability to perform predetermined functions for bridges.And it can be predicted the structural performance trends in the future through time-dependent reliability.However,the researches about bridge reliability are not fully applicable to existing medium and small span bridges in seasonal frozen region,of which the adverse environmental conditions are not considered perfectly.Therefore,it is of great significance to accurately and effectively study the reliability of existing medium and small span bridges.This study is founded by the National Natural Science Foundation Project "Research on modal parameter identification and damage detection of bridge structure considering temperature effect" and Science and Technology Project of Department of Transportation of Jilin Province "Research on safety evaluation and maintenance strategy of expressway bridges in seasonal frozen region based on time-dependent reliability".In this paper,time-dependent reliability of existing medium and small span bridges in seasonal frozen region has been studied with the main influence factors indentified.The reliability calculation method for medium and small bridges is proposed considering the influence factors,such as chloride ion erosion and freezing-thawing cycles.The bridge deflection monitoring method is established based on modal flexibility theory and Kriging method improved by artificial bee colony algorithm.The serviceability reliability of bridge is computed and analyzed by dynamic Bayesian network.And it is studied that the damage identification method for medium and small span bridges.The specific researches are described as follows:1.Considering chloride ion erosion influenced by freezing-thawing cycles,the rusted corrosion degree and residual tensile strength time-dependent models for steel bars and prestressed steel strands are proposed under the combined action of chloride ion erosion,freezing-thawing cycles and environment temperature effects.The time-dependent model of concrete compressive strength is proposed,which is influenced by freezing-thawing cycles.As for load action,it is simulated that the uncertainty of dead load and the randomness of traffic.Therefore,the time-dependent reliability calculation method for the bending capacity of medium and small span bridges is proposed based on the mentioned models.A simple supported hinged slab bridge and a three-span prestressed continuous girder bridge in Jilin Province are used as representative of typical small and medium span bridges.The bridge bending resistance is evaluated through the time-dependent reliability and the effect of influence factors on the bridge reliability is analyzed by sensitivity method.2.The hinged joint damage of the simply supported hinged slab bridge and the board and girder damage are taken as the main damage forms of medium and small span bridges.The eccentric load coefficient and overload coefficient of the vehicle load are studied to reflect the disadvantageous live load action on the bridge.The relative displacement between slabs is regarded as the damage indicator of modified hinge plate method and the bridge bearing capacity reduction coefficient is used to evaluate the board and beam damage.And it has been studied that the effect of unfavorable load and damage state on the reliability of the bridge.3.A bridge deflection calculation method is proposed based on modal flexibility theory.The mode shape of bridge is computed by Kriging method used as surrogate model so as to achieve that the bridge deflection at arbitrary measure point can be calculated through a small amount of mode shape node.In order to improve the calculation accuracy,Kriging method is optimized by artificial bee colony algorithm.Through the field test,the feasibility and accuracy of the proposed deflection calculation method is verified under the comparison between traditional static test and the proposed method in this paper.4.Combining with environmental influence factors and bridge dynamic characteristic health monitoring,the serviceability reliability of medium and small span bridge in seasonal frozen region is calculated and analyzed with bridge deflection chosen as the evaluation indicator of bridge serviceability limit states.A dynamic Bayesian network is employed to establish the relationship between environmental temperature,humidity,modal frequency and mode shape.The probabilistic distribution model of the influence factors is updated by monitoring data,which improves the accuracy of bridge deflection reliability assessment.It is analyzed that the relationship between reliability index and influence factor.The effects of monitoring factors on bridge deflection reliability are analyzed by sensitivity parameter.5.It is studied that the damage identification method for the common forms of simply supported hinged slab bridge and continuous bridges.A damage identification algorithm for hinge joint damage of simply supported hinged slab bridge is proposed based on modified hinge plate method and swarm intelligence algorithms with lateral load distribution as the damage indicator.It is studied and analyzed that the applicability and effectiveness of original artificial bee colony algorithm,improved artificial bee colony algorithm,genetic algorithm and particle swarm optimization for the proposed damage identification method.The modal curvature difference is used as the beam damage identification indicator.With the aim of damage identification without relying on the intact structural information,Kriging method is employed to reconstruct the actual measured modal curvature so as to get the modal curvature of intact structure.The proposed damage identification methods are verified by simulative damage conditions of simply supported hinge slab bridge and the three-span continuous beam bridge.