Research On Bridge Damage Identification Based On Vehicle Scanning Method And Empirical Wavelet Transform

With the rapid development of infrastructure construction in our country,a large number of small and medium-span bridges have been in service for more than 20 years.Due to the rapid growth of traffic volume and the long-term existence of overloading,the health of small and medium-span bridges is facing a severe challenge.At present,the health condition assessment of small and medium-span bridges based on manual inspection has the disadvantage of being time-consuming and labor-intensive.To this end,Yang et al.proposed the Vehicle Scanning Method(VSM).They measured the acceleration response of the vehicle that installed an acceleration sensor during driving on the bridge.Then they used acceleration response to identify the modal parameters of the bridge and further identify bridge damage based on the vehicle-bridge coupling analysis.The bridge damage identification method based on VSM is fast,efficient,and saves manpower and material resources.This paper proposed a new bridge damage identification method based on VSM.Firstly,the vehicle acceleration response and the contact point acceleration response were obtained after the vehicle passed through the undamaged and damaged bridge.Secondly,the intrinsic mode function(IMF)corresponding to the driving frequency was obtained by using the Empirical Wavelet Transform(EWT)to process the acceleration response.Finally,the difference of the IMF was used to construct the damage indicator to identify the bridge damage.This paper will use vehicle acceleration response and contact point acceleration response to identify the damage.The main research contents and conclusions of the paper are as follows:Firstly,the theoretical basis of bridge damage identification based on VSM was introduced,taking the single-axle vehicle model as an example.The theoretical expressions of the bridge,vehicle,and contact point responses of the single-axle vehicle-bridge coupled vibration model were induced.The frequency components of the contact point responses were analyzed.The principle of EWT and a new damage indicator were introduced.The results show that the contact point response only includes driving frequency and bridge frequency components but no vehicle frequency components.So we can avoid the interference of vehicle frequency on damage identification by using the contact point response.Secondly,the proposed damage indicator was used to identify the bridge damage based on the single-axle vehicle model.The response of the single-axle vehicle-bridge coupled vibration system was calculated.The ability of the proposed damage indicator to identify single damage,multiple damages,and minor damages of simply supported girder bridges were analyzed.Then,the effects of bridge damping,vehicle damping,vehicle-bridge mass ratio,vehicle speed,measurement noise,and road surface roughness on damage identification were analyzed.The results show that the damage identification results were consistent based on the vehicle response and the contact point response.The constructed damage indicator can identify single and multiple damages at different locations.Different damage locations have different peak values of the corresponding damage indicator.The damage indicator is highly sensitive and can identify 1% of minor damage.The bridge and vehicle damping have a certain influence on the damage identification results.The smaller the test vehicle damping is,the better the damage indicator identification results are.For different vehicle-bridge mass ratios,the closer the vehicle frequency is to the bridge’s natural vibration frequency,the better the damage identification results is.We all know that the actual vehicle frequency is higher than the bridge’s fundamental frequency.Therefore,better identification results can be obtained by increasing the vehicle mass in the actual test.Under the condition of 20% level measurement noise and A level road surface roughness,the proposed damage indicator can still clearly identify the damage location.It indicated that the damage indicator has certain noise immunity.Then,the vehicle model was extended to the two-axle vehicle.The ability of the proposed damage indicator to identify the damage of the simply supported beam was further investigated.The damage identification results of the two-axle vehicle were compared with the single-axle vehicle.The results show that the proposed damage indicator can identify the multi-damage and minor damage of the simply supported beam based on the two-axle vehicle-bridge coupled vibration model.And the damage identification results are promising.The influence of factors such as vehicle-bridge mass ratio,vehicle speed,measurement noise,and road surface roughness on the damage identification results of two-axle vehicles is the same as that of single-axle vehicles.It should be noted that the filtered range of the driving frequency affects damage identification accuracy due to the front and rear axles of two-axle vehicles on and off the bridge.If the filtering range was selected larger,the damage location would be more easily exposed,and the identification result will be more accurate.However,if the filtering range is selected too large,more interference will be introduced.Therefore,in order to obtain the best identification effect,it is necessary to select and adjust the filtering range.Finally,the damage identification model experiment of simply supported girder bridge was designed to verify the feasibility of the proposed damage indicator in practical application.The experimental results show that the natural vibration frequency of the bridge-vehicle system can be accurately identified based on the VSM.We were able to determine the existence of damage based on the change in the natural vibration frequency.By increasing the mass and speed of the test vehicle,the amplitude of the driving frequency can be increased,and the visibility of the driving frequency can be increased.The proposed damage indicator can successfully identify the single damage and multiple damages of the simply supported beam.However,the vehicle’s local vibration and measurement noise itself seriously interfere with the damage identification results.