Dynamic Detection And Diagnosis Methods For Beam Bridge Based On Vehicle Excited Vibration Response

Because of the deterioration of material performance and damage caused by external load,the existing bridges will face security risks after a certain years of service.A bridge service status detection and diagnosis method is the key point to ensure bridge safety.The detection methods that based on bridge load test have convenient operation mode and good detection ability of bridge structural status.Based on the current engineering and technical status,bridge detection methods based on load test have a wider space of development compare with other detection methods.According to the current standard guidelines and engineering practice status,most load tests give priority to static load test,and the dynamic load test often play an auxiliary role.The dynamic load test is usually under-utilized.However,the results of dynamic load test contain all-side structural information.They can present more detailed bridge stress and damage conditions.Therefore,it is necessary to take a further study on the dynamic detection branch of bridge load test.Take the beam bridge dynamic response data excited by moving vehicle load as analysis object,the following research contents about bridge condition diagnose method based on dynamic load test will be conducted in this dissertation:The effects of nonstructural influence factors of IM are analyzed.First,the influence of pavement roughness condition on IM is separately analyzed in spatial domain and frequency domain.Second,the varying trends of IM change with vehicle velocity and weight are analyzed.The results show that the influences of vehicle factors depend on the pavement roughness conditions.They will affect the values of IM jointly.Third,how the spaces between vehicles for a vehicle team load affect the IM is studied.The GA optimization algorithm is used in the analysis.The results show that vehicle space has different effect rule for different space range.Thus,the varying trend of IM should be treated differently for different vehicle team load arrangement.The limitation of current pavement roughness grade method by PSD is proved by both theoretical analysis and numerical simulation.The probabilistic model of IMs that belong to one same PSD grade is established.And a new index called feature impact factor ?fe is proposed based on the probabilistic model.The new index can represent the relationship between roughness grade and bridge dynamic response increment in different assurance rate,which will improve the representation of pavement roughness PSD grade on IM.Furthermore,a novel IM detection method considering the deterioration of pavement roughness is presented.The method simulates the deterioration of pavement by place obstacles on the pavement.And the objective incensement of IM is set equivalent to the feature impact factor.The evaluation suggestions are also proposed based on different testing results.A new IM detection and prediction methods considering vehicle team load is also presented.This detection method can reduce the gap between bridge testing load condition and bridge service load condition.The combination of real testing results excited by one vehicle and GA search algorithm is considered in the method.By the use of this new method,the difficulties of field dynamic test by multi vehicles can be avoided,and more distribution patterns of vehicle team load can be considered.The most unfavorable IM excited by vehicle team load can be seeked out by this detection method.Studies on the signal processing and parameter identification methods of dynamic detection signal are conducted.First,a denoising program based on wavelet coefficient correlation is composed.The program improves the retention of high frequency details of signal after denoising.Second,present three methods which are able to separate the dynamic and static components of a dynamic displacement curvel,such as curve fitting method by least square,low pass filtering method by the comparison of amplitude spectrum,and EMD method.These three methods are mutually independent.Thus,they can be solely used,or joint used in order to improve the accuracy and reliability of the calculation result of IM.Third,a dense modal parameters identification method based on the optimal complex Morlet wavelet analysis is presented.By means of parameter optimization of the Morlet wavelet equation,the frequency domain resolution can be enhanced without recording more testing signal.Based on the moving vehicle excited test which is commonly used in bridge dynamic detection,a damage detection method based on continuous wavelet transformation is proposed.The dynamic displacement signal is selected as the source data.The method locates the position of damage by the distribution of module maximum traces and quantified the damage degree by Lipschitz index.At the same time,the influence of variable factors on the effect of damage detection,such as pavement conditions,vehicle load parameters,and measurement position,are all analyzed.Some practical suggestions about enhancing the accuracy of damage dictation are also proposed.