Study On Damage And Vehicle Load Identification Method Of Small-and Meddle-span Girder Bridges Based On Distributed Sensing

With the rapid development of transportation in China,the highway mileage has reached more than 161,000 kilometers.Bridge plays an important role in highway.More than 85%bridges in China are small-and medium-span bridges.However,due to the low maintenance frequency and the lack of health monitoring system,there have been different kinks of defects on them.It is necessary to develop effective health monitoring methods for small and mediumspan bridges.Considering that unknown damages and vehicle loads can jointly affect the degradation of bridge.Damage identification,random traffic flow parameter identification,and synchronous identification methods are developed in this study,and taking advantages of long-gauge fibre Bragg grating sensors.The contents and achievements are given as follows:(1)Rapid damage detection method based on distributed sensing and fractal dimension theoryTaking simply supported bridge as the research object,fractal dimension curves under moving vehicle are calculated based on the strain history recorded by long-gauge strain sensors,and the fractal dimension curves are normalized to eliminate the influence of vehicle parameters.Bridge damages can be located by the change of the normalized fractal dimension curve before and after the damage.The proposed damage index is sensitive to bridge damage even in multiple damage cases,but not relevant to vehicle parameters.(2)Reference-free bridge damage identification method based on distributed sensingTaking continuous beam as the research object,the strain history equation of the continuous bridge under moving vehicles is deduced,then the relationship between the strain history integral curve and section number are derivated.Bridge damages are located by the abnormal point of the first-order difference curve.Bridge damage extents can be calculated with the relationship between the strain history integral of the damaged bridge and the extents of damages.The proposed method is verified with an indoor experiment and a numerical simulation.The indoor experimental results show that the proposed method is effective under variable vehicle parameters,and bridge damages can be located even damages happen in multiple span of bridge.The identification error of damage degree is up to 50% due to the influences of experimental error,and the identification accuracy increases with the degree of damage.In addition,the results of being insensitive to road surface conditions and applicable for damage detection of multi-span continuous bridge can be concluded from the numerical simulation,and the damage degree identification error is less than 10%.(3)Random traffic flow parameters identification method based on distributed sensingPassing a vehicle with known parameters through bridge to calibrate bridge strain influence line.The peaks of the second-order strain difference curve of three adjacent sensors are used to determine axle lane,wheelbase,speed,and distance between vehicles.Axle weight are calculated with the calibrated influence line and bridge weigh-in-motion system.The proposed method can realize the identification of vehicle direction,lane,wheelbase,distance,speed,and axle weight.A three-dimensional bridge-vehicle coupling system is established,the accuracy and effectiveness of the proposed method is verified under varying road surface roughness,vehicle acceleration and number of vehicles.The results of indoor experiment tallies with the results of bridge-vehicle coupling system.Besides,the effectiveness of the method under both direction traffic flow also has been verified.(4)Theory of simultaneous identification of bridge damage and dynamic vehicle loadBridge dynamic equation is adopted to establish the objective function,bridge damages and vehicle parameters are regarded as parameters to be solved.Bridge damage locations are incorporated into the iterative parameter optimization process of vehicle and bridge parameter identification,to reduce unknown damage parameters.Vehicle wheelbase,vehicle spacing,axle dynamic load of multi-vehicles,and bridge damage location,bridge damage degree are simultaneously calculated through hierarchical Bayesian iterative optimization.The effectiveness of the proposed method under increased noise,variable road roughness,multiple damages,and varying damage levels are verified.