Study On Method Of Damage Identification Based On The Hilbert-huang Theory For Stay-cables

Because of advantages which the cable-stayed bridge has, it is popular with engineers and people in the construction of modern bridge. This paper makes full use of the predecessors’ research results in the field of signal analysis and the cable-stayed bridge in actual service, puts forward the difficult problems to solve the Hilbert- Huang transform, and compared with wavelet analysis. By establishing a finite element model, study on the cable damage identification of cable-stayed bridge based on Hilbert-Huang transform by the numbers.The main contents of this paper include:Firstly, by analyzing the Hilbert-Huang transform theory, the termination criterion of EMD is given, and analyzes its theory which is not mature. Compared with the wavelet analysis method, this paper gives its advantages in the damage identification, and cited examples prove that Hilbert-Huang transform is more simple and effective.Secondly, it is lack of stability that based on the original energy curve to distinguish the cut-off point of noise and signal, and according to the different statistical characteristics of signal and noise, a new EMD de-noising method is presented based on autocorrelation function characteristics. Through the experiment used the de-noising method compared with Fourier transform filtering de-noising method and wavelet transform de-noising method, validate the effectiveness of the proposed method.Once again, the mechanism of end effect and the existing processing methods are summarized, an adaptive endpoint sine phase continuation method is proposed on its basis(only the EMD end effect processing). The simulation analysis shows that, the method can not only solve the end effect of the signal, but also compared to existing continuation methods, the number of screening will be less, significantly improve the accuracy of signal EMD decomposition, thereby reducing the end effects of Hilbert spectrum.Finally, the modified Hilbert-Huang transform method combines with finite element analysis models, for damage identification analysis and discussions of the cable of cable-stayed bridge in service. Through the experiment under all the damage conditions, can effectively identify the damage location and degree of the cable.In conclusion, this paper based on the signal analysis and processing method of Hilbert-Huang transform, for the given termination criterion of EMD, the proposed methods based on autocorrelation function characteristics of the EMD de-noising and the adaptive endpoint sine phase continuation are used, combines with the finite element model, the accurate damage location and recognition of degree for the cable of a cable-stayed bridge in service is realized.