Damage Identification Of Simply Supported Concrete Girder Bridge Based On Dynamic And Acoustic Characteristics

Reinforced concrete(RC)girder bridge is one of the most widely used bridge types.Under the coupling action of environment and load,steel and concrete are susceptible to damage and deterioration,which seriously affect cthe service performance of structure.It leads to the structure not meeting the normal service performance or bearing capacity requirements,or even collapse of bridge.Therefore,it is essecnital to identify the damage status of RC girder bridge in time and formulate reasonable maintenance strategy to ensure the safe operation of bridge structure.Damage identification and characterization based on dynamic and acoustic characteristics have been widely used in the field of bridge health monitoring.Vibration-based damage identification methods usually rely on modal data of baseline structures as a benchmark.Most damaged bridges were built many years ago,and the acquisition of their basic material properties(elastic modulus,density and damping ratio),geometric characteristics(cross-sectional size,etc.)and support conditions are complex and even impossible.Therefore,modal-based damage identification method independent of reference model presents good application prospect.Acoustic emission(AE)is a local,high sensitivity and passive nondestructive testing method.It is essential for the practical application of this technology in bridge structures to clarify the correlation between acoustic characteristics and static,dynamic characteristics of reinforced concrete beam bridges in the process of fracture and to realize the fracture characterization of based on the pattern recognition of AE parameter.This thesis takes the most widely used reinforced concrete beam bridges as research objects.On the one hand,a damage identification method without considering baseline model data is proposed based on structural modal parameters.On the other hand,structural damage and fracture processes are characterized based on structural acoustic characteristics.The specific research contents carried out in this thesis include:(1)The damage identification accuracies of modal shape curvature and uniform load surface curvature indexes were verified by setting the damage identification conditions in unit position and multi-location for simply supported girder bridges.Considering the influence of noise,number of sensors,modal order and other factors,the influence analyses of modal shape curvature index and uniform load surface index were carried out,which provided a basis for selecting damage identification index of simply supported beam bridge with multiple girders.(2)Aiming at the difficult to obtain the baseline model of reinforced concrete beam bridge,a damage identification method without considering baseline model data was proposed based on Chebyshev polynomial and neural network fitting combined with fuzzy reasoning theory.Firstly,the modal shape curvature data of baseline model were obtained by Chebyshev polynomial and neural network fitting method based on the modal shape curvature curves of the damaged multi-girder simply supported beam bridge.Corresponding modal shape curvature difference was obtained by calculating the modal shape curvature values of the post-damage model and the baseline one,and the structural damage location was identified.The applicabilities of Chebyshev polynomial fitting and neural network fitting methods were also compared and analyzed.Finally,the modal shape curvature difference was used as the input parameter of the fuzzy inference system to identify the damage degree of multi-slice simply supported beam bridge.(3)Through the four-point bending test of RC beams,structural static,dynamic and acoustic parameters were obtained.Damage and fracture characterization of RC beams was realized,and the correlations among static,dynamic and acoustic parameters were clarified.Acoustic emission parameters(hit,amplitude,energy,count,duration)and their cumulative variation trends were consistent with the crack generation and propagation patterns,which can be used to characterize the damage and fracture of structures.Damage indexes based on acoustic emission energy and hit showed different trends with the increase of load,which can effectively identify the density and damage degree of structural cracks.Accumulative acoustic emission parameters were positively correlated with the cumulative crack width of beams.The first and second order modal frequency variation rates increased with the increase of load level and cumulative crack width,which showed favorable correlation.(4)Based on the statistical characteristics of acoustic emission parameters,a characterization method for damage and fracture of reinforced concrete beam bridges was proposed.Firstly,data compression and information mining were realized based on principal component analysis(PCA)of acoustic emission parameters(hit,amplitude,energy,ringing count,duration and rising time),and the representation objects of each principal component were defined.Secondly,damage location of the structure was determined based on the extreme difference analysis and variation coefficient analysis of b-value data.Finally,damage and fracture process and fracture mode were identified based on the Gaussian filtering analysis of acoustic emission energy and b-value data.Based on the statistical characteristics of acoustic emission parameters,the damage characterization method of RC beam bridge can effectively improve the interpretation ability of parameters.The research results of this thesis provide a new method for damage identification of RC beam bridge,and present important theoretical value and practical significance to ensure the safe operation of bridge.