Study On The Signal Recognition Method Of Bridge Cable Break Based On Acoustic Emission Technology

In recent years,with the continuous acceleration of basic transportation construction,the demand for intelligent transportation technology has become increasingly widespread.As an important transportation system,the safety of bridges is a major issue related to national economy and people's livelihood Today's long-span bridges are mainly cable-stayed bridges and suspension bridges with cable-bearing systems,the main stress component of this type of bridge is the cable.When the cable is used continuously for a long time,it is easy to be fatigued or corroded,which causes the tensile strength to decrease.Even worse the broken cable may cause the bridge to collapse,therefore,long-term real-time monitoring of bridge cables is of great significance.Acoustic emission technology,as a non-destructive real-time monitoring technology,has been proven to be an effective method for monitoring the dynamic damage of materials.In this thesis,the prestressed steel strand material commonly used in bridge cable construction is taken as the research object.Oil pumps are used to load prestressed steel strands at a uniform speed for tensile fracture tests.At the same time,the acoustic emission equipment is used to collect real-time signal from the steel strand during the tensile test.Based on the typical acoustic emission signal processing technology,the typical acoustic emission characteristics of the prestressed steel strand broken wire source have been analyzed.A method for identifying the acoustic emission signal of the broken wire has been proposed.The main contents and results of this thesis are as follows:(1)Through the tensile test of prestressed steel strand material loaded at a uniform speed to the occurrence of a broken wire event,the acoustic emission signal of the entire process is obtained.The classical acoustic emission signal simplified waveform feature extraction method is used to calculate the time domain waveform characteristic parameters of the acoustic emission signal.The Fast Fourier Transform method is used to calculate the frequency domain characteristics of the acoustic emission signal.Then,the loading time is plotted on the abscissa,the correlation diagram of the characteristic parameters of the acoustic emission signal and the mechanical parameters of the steel strand were drew,and the change trend of each acoustic emission characteristic parameter with the loading process was summarized.The characteristic parameters of the broken wire acoustic emission signal are used to construct the feature vectors for distinguishing different types of acoustic emission signals(2)According to the characteristics of the feature vector,a method for judging the occurrence of a broken wire in a cable is studied.Based on the principal component analysis method,a model of the characteristic parameters of the acoustic emission signal when the wire is not broken is established.The control limits of the normal model T2 statistic and SPE statistic are calculated based on the kernel density estimation.When applied to actual monitoring,the feature vector of the acoustic emission signal collected in real time is mapped to the low-dimensional space according to the load matrix of the principal component analysis,and then T2 statistics and statistics are calculated When the two statistic values exceed the control limits of the normal model T2 statistic and SPE statistic,it shows that the signal does not conform to the normal situation model,so that the broken wire signal can be identified.The experimented results in this thesis show the effectiveness of the method.