Multi-vehicle Dynamic Load Of Multi-girder Highway Bridge: Identification Method And Experimental Investigation

With the rapid development of traffic investigation,over-limit management and charging by weight in recent years,bridge weigh-in-motion(B-WIM)systems have been more and more widely used.The application of road type B-WIM for vehicle axle detection is affected by the adverse factors during installation and use.Therefore,the non-pavement B-WIM based on the bridge dynamic response has been extensively developed for vehicle axle detection.However,the application scope of non-pavement B-WIM system is limited as it is effective when single vehicle crossing the bridge only.The identification method for the multi-vehicle dynamic load of multi-girder highway bridge by using the B-WIM system is investigated in this paper.Firstly the dynamic response of bridges generated by passing vehicles is divided into local effect and global effect.Then the local effect and global effect are employed for vehicle axle detection and axle load and total weight identification,receptively.Theoretical analysis,numerical simulation and experimental studies are conducted for the identifications of vehicle axle,axle load,and total weight more effectively and practically.The main work and conclusions are as follows:1.Based on the theory of the vehicle-bridge coupled vibration,the models of bridge and vehicle are developed for simulating the situation of multi-vehicle crossing the bridge.The equation of vehicle-bridge coupled vibration is derived accordingly.The road roughness and the bridge motion state are used as the excitation source of the vehicle system.Thus each vehicle moves according to its respective excitation when conducting response analysis of multi-vehicle.The contact forces at all the time points of all the wheels of the multi-vehicle are calculated by Matlab based on the excitation source,and the dynamic response of the bridge is calculated by the ANSYS.The combined utilization of the self-made program and commercial finite element software is suitable for the vibration analysis of complex bridges.2.The identification algorithm based on influence line fitting technique is proposed for non-pavement B-WIM.Considering the lateral distribution of the wheel loadand the two-dimensional stress of the bridge,the Moses algorithm using the stress influence line and the axle information is extended from one-dimensional to two-dimensional.The numerical simulation results indicate that the proposed method has obvious advantages compared with the method without considering the lateral distribution of the wheel load.3.The practical feasibility of multi-vehicle load identification algorithm is investigated via field test.The transverse load distribution coefficient is obtained by using the strain response measured via the calibrated vehicle.The accuracy of the proposed influence line and load identification algorithm is verified via the field test.Correction factors are introduced to modify the vehicle load identification equation for the bridge with small transverse load distribution coefficients when necessary.4.The non-pavement B-WIM method to estimate the axle weight of multi-vehicle highway bridge in which strain signal of single vehicle is obtained by separating the multi-vehicle mixed signal is presented via blind signal separation(BSS).The mixed signal model of non-pavement multi-vehicle B-WIM system is established based on the superposition principle of elastic structure.The signal preprocessing method and the separation steps of multi-vehicle signal are introduced.The effectiveness of the BSS based algorithm is verified by the experiment in the laboratory.The results indicate that the multi-vehicle problem can be transformed into single vehicle problem with the BSS,which provides a new way to identify the multi-vehicle load in non-pavement B-WIM system5.The test accuracy evaluation method for non-pavement multi-vehicle B-WIM and the standard of the accuracy grade evaluation are introduced.The test accuracy analysis is carried out in the aspects of applicability;temperature effect,sensor test and vehicle speed,and the corresponding strategies for improving accuracy and long-term stability of non-pavement B-WIM are proposed.