Method for identifying structural damage based on vibration transmissibility function and support vector machine

Abstract

The invention relates to the field of structural damage identification, in particular to a method for identifying structural damage based on a vibration transmissibility function and a support vector machine. The vibration transmissibility function is constructed through an accelerated speed response, the amplitude value of the vibration transmissibility function is used as an analysis signal, all frequency band signals are obtained through wavelet packet decomposition and reconstruction, the energy ENj of all the frequency band signals is obtained through calculation, the variable quantity delta ENj of the energy of the front and rear frequency band signals of the structural damage is used as damage characteristic indexes, and the mode identification function of the support vector machine is used for identifying the structural damage. The method for identifying the structural damage is not influenced by the excitation amplitude value, meanwhile, a structural damage identification method with the problems that a neural network structure is hard to determine and prone to being in local minimization is avoided, operability is strong, processing precision is improved, and the damage identification process is simplified.

Description

technical field [0001] The invention relates to the field of structural damage identification, in particular to a structural damage identification method based on a vibration transmissibility function and a support vector machine. Background technique [0002] Due to the effects of various loads and environmental factors, civil engineering structures will be damaged to varying degrees during service. If they are not detected and taken measures in time, the consequences will be unimaginable. Therefore, it is necessary to carry out damage identification research on structures. So far, many structural damage identification methods have been researched and proposed, among which, the structural damage identification method based on vibration response analysis is a promising method. [0003] At present, the acceleration response of the structure is directly used as the analysis signal for wavelet packet decomposition and reconstruction, and the energy of each frequency band is cal...

AI Technical Summary

Problems solved by technology

[0003] At present, the acceleration response of the structure is directly used as the analysis signal for wavelet packet decomposition and reconstruction, and the energy of each frequency band is calculated. The energy difference of each frequency band before and after the structural damage is used as the structural damage feature vector, and when the structural damage is identified through the neural network, It will be affected by the excitation amplitude, and at the same time, it will encounter problems such as difficulty in determining the structure of the neural network and easy to fall into local minima.

Images