Structural damage early warning method based on emd decomposition and ar model residuals

Abstract

The invention relates to the field of structural damage identification and provides an EMD decomposition and AR model residual error based structural damage early-warning method According to the method, an EMD is utilized to decompose actually-measured acceleration response signals into a plurality of intrinsic mode functions containing single frequency components, then an AR model is utilized to conduct fitting on two large-energy intrinsic mode functions IMF, residual errors of the AR model are extracted, an elliptic control equation is established according to variances of the residual errors of the AR model, and structural damage early-warning is performed by utilizing the elliptic control equation. Compared with an existing structural damage identification method based on vibration, the structural damage early-warning method directly utilizes nonlinear and non-stationary acceleration response signals to perform structural damage early-warning, does not need a standard finite element model and modal parameters, is not influenced by uncertainty factors, such as nonlinear and non-stationary actually measured response signals, incomplete modal parameters and measurement noise, belongs to a data-driven method and is high in calculation efficiency, good in operability and wider application range.

Description

technical field [0001] The invention relates to the field of structural damage identification, in particular to a civil engineering structure damage early warning method based on vibration response signal processing technology. Background technique [0002] Modern civil engineering structures continue to show the trend of large-scale, complex and diversified. Due to the impact of environmental erosion, material aging and load effects, civil engineering structures will be damaged to varying degrees during the service period. If the damage is not detected and dealt with in time, once the damage of some key components of the structure accumulates to a certain would expand rapidly, compromising the security of the entire structure, with catastrophic consequences. Therefore, it is necessary to conduct structural damage identification and health monitoring research. [0003] In recent years, the vibration-based structural damage identification method has developed rapidly. Since...

AI Technical Summary

Problems solved by technology

[0004] In view of the problem that the existing vibration-based structural damage identification method is affected by the nonlinearity and non-stationarity of the response signal during the application process, resulting in unreliable results, a structural damage early warning method based on EMD decomposition and AR model residual is proposed In this method, the measured acceleration response signal is decomposed by EMD which is suitable for dealing with nonlinear and non-stationary signals, and the AR model is used to fit the intrinsic mode function IMF with large energy, and the residual of the AR model is extracted. The variance of the ellipse control equation is constructed, and the structural damage warning is carried out by using the ellipse control equation constructed. The whole process does not require a benchmark finite element model and modal parameters, and the results are reliable and operable.

Images