Acoustic Emission Signal Processing And Damage Analysis For Bridge Cable Fatigue Damage

Cable-stayed bridge as an important force component plays an important role in the cable stayed bridge. However, they are often suffered from damage due to corrosion,fatigue and external shock, and these damage gradually evolving and not avoid to damage accumulation or resistance decay. At extreme cases, the catastrophic accident happened. To ensure the cable stayed bridge normal operation, the best mothod is to monitor its health condition. Acoustic emission(AE) is an effective method to monitor the bridge cable damage. Because AE signals susceptible to external environmental interference, so the signal analysis method is adopted to process the AE signal, to effectively reveal the status of cable damage. Therefore, the appropriate signal processing method is used to analyse the AE signal, and to makde the reasonable evaluation for the calbe damage situation, to ensure bridge safety in service time.The main contents of this paper, as follows:(1) The wavelet singularity detection theory is used to describe the different damage stages for bridge cable, and the effectiveness of this detection method is proved by the matlab simulation signals. Then, the method for combine the 3cr criteria and hard threshold is adopted, and analyse the effect of this method in dealing with noise AE signal. Last, the wavelet packet is used to decompose the AE signal, and adopt the energy of each band to construct the features vector, and calculate the Euclidean distance for eack feature vector to identify the different types of signals.(2) The fatigue damage process for the carbon fiber cable and the multi-age steel cable is studied, respectively. The specifically content includes:(a) the noise effectively remove and the AE signals is effectively isolated from the noise environment, (b) the index of the Kurtosis and the b is used to describe the damage evolution process of carbon fiber cable, and the AE parameters method combined with RA value to analyse the characteristics of each stages, (c) the features waveforms are extracted, and the AE modal analysis technology is adopted to analyse the mechanism of sound source waveform generator and the frequency range of every damage waveforms, (d) the singularity detection combined with the AE parameters method to study the fatigue damage evolution process of multi-age cable, the degree of damage and the fatigue damage phases, (e) the fault detection factors are used to identify the broken wires of multi-age cable, and to analyse the mechanism for damage sound source of corrosion steel cables. (3) The detection method of cable fatigue damge based on the principal component analysis is adopted. The specifically content includes:(a) noise impact on the principal component model, (b) the damage detection accuracy of principal component analysis is studied under noise interference, (c) the original signal characteristics is maximize highlighted by the signal fusion technology of the multi-channel signal and the different AE parameters. (d) the multi-channel signals is fused by the principal component fusion method, and this method is more effective and precise compared with traditional weighted fusion method, and based on the fatigue damage data of multi-age bridge cable, the positioning accuracy and the pattern recognition for fusion signal are improved.