Variable cross-section beam damage identification method based on node curvature and wavelet analysis

Abstract

The invention provides a variable cross-section beam damage identification method based on node curvature and wavelet analysis. The method is based on the assumption that structure microdamage has a little effect on stress redistribution of a variable cross-section beam. At first, displacement signals generated when concentrated loads pass through a damaged variable cross-section beam at a constant speed are measured, node curvature matrixes generated before and after the variable cross-section beam is damaged are calculated based on an undamaged single-scale finite element model, damage location identification is achieved, the structure damage degree is precisely calculated on the basis of a space-time multi-scale model, and finally the damage condition of the variable cross-section beam is obtained.

Description

technical field [0001] The invention belongs to the field of civil engineering, and relates to a damage identification method for variable-section beams based on node curvature and wavelet analysis. Background technique [0002] With the progress of society and the development of civil engineering technology, the health monitoring and safety status assessment of large civil engineering structures, especially large bridge structures, has become a hot spot in the academic and engineering circles at home and abroad. How to invert the working state and health status of the bridge structure according to the collected data and signals, establish an accurate finite element model, accurately identify the structural damage location and its damage degree, and serve the damage and safety prognosis of the bridge structure. It is one of the most core and critical issues in health monitoring, injury diagnosis and safety assessment. [0003] At present, the research on damage identificati...

AI Technical Summary

Problems solved by technology

[0003] At present, the research on damage identification methods at home and abroad mainly focuses on model-based and data-driven. The data-driven damage identification method is based on the massive time history collected by the structural health monitoring system. Data, the damage is identified through the nonlinear characteristics of the structural unit derived from the data itself or related changes. The main methods include sequential model method, wavelet analysis method, Hilbert-Huang transform (HHT), Kalmanfilter method, etc. This method can quickly identify the occurrence of damage time and location, and can even achieve real-time damage identification, which has a positive effect on the real-time early warning based on the online health monitoring system; while the model-based damage identification method mainly refers to using the measured information of the structure to continuously modify the finite element model. Structural damage identification, this method can understand the monitored structure as a whole, and can effectively predict, but the amount of calculation and data storage is large, often requires very cumbersome calculation processes and programs, and it is almost impossible to achieve real-time predict

Images