Accurate Extraction And Application Of Quasi-static Components From Fast Moving Vehicle-induced Dynamic Response

Tens of thousands of existing highway bridges in China are subjected to heavy traffic loads,and fast moving vehicles are the most important dynamic loads on bridges.It has been shown that the fast moving vehicle-induced dynamic response signal consists of quasi-static components and dynamic components caused by the bridge natural frequency,among which the quasi-static components are more sensitive to bridge damage and less sensitive to environmental changes(e.g.,temperature),showing good prospects for application in several fields.Accurate extraction of quasi-static components from the dynamic response is a prerequisite and key for subsequent applications,but it is necessary and challenging to accomplish this task considering the different vibration characteristics of different bridges,road surface roughness,and speed of moving vehicles.The purpose of this thesis is to study the accurate extraction of quasi-static components based on the fast moving vehicle-induced dynamic response and its application,covering the following three aspects:(1)to study the effects of road surface roughness,vehicular speed and other factors on the fast moving vehicle-induced dynamic response;(2)to study the adaptive and accurate extraction of quasi-static components based on the fast moving vehicle-induced dynamic response;(3)to study the damage identification and calculation methods of dynamic amplification factor based on the quasi-static components.Firstly,the analytical solution of the vehicular dynamic response of a simply supported beam is derived by considering the road surface roughness and it is found that the vehicle-induced dynamic response of the bridge contains two components,namely,the quasi-static component and the vibration component caused by the bridge natural frequency and road surface roughness excitation.The quasi-static component is determined to be a good reflection of the inherent characteristics of the bridge and the key problem of extracting the quasi-static component from the dynamic response of the bridge is analyzed:Then,the thesis proposes an improved variational mode decomposition(VMD)algorithm to extract the quasi-static components of the fast moving vehicle-induced dynamic response through the analysis of the dynamic response components.The method consists of two key steps:first,the adaptive determination of two key parameters of the VMD algorithm based on the particle swarm algorithm(PSO);second,the improvement of the endpoint effect of the VMD algorithm through the mutual information entropy criterion.Then,the feasibility of the quasi-static adaptive extraction method proposed in this paper is demonstrated in terms of accuracy,robustness and applicability by numerical examples and practical engineering validation,taking into account the grade of road surface roughness,vehicular speed,noise,different bridge types and other factors.Finally,by deriving the analytical solution of the vehicle-induced dynamic due to the damage state of bridge,it is determined that the dynamic response of the bridge after the damage contains two signal components,i.e.,the quasi-static component and the vibration component caused by the bridge natural frequency and the excitation of the road surface roughness.Then,we analyzed the characteristics of the two signal components after the damage,and found that the quasi-static component has better sensitivity to the damage.The quasi-static components are used in the bridge damage identification and the calculation of dynamic amplification factor and good results are achieved.