Structural damage identification method based on echo state and multi-scale convolution joint model

Abstract

The invention discloses a structural damage identification method based on the joint model of echo state and multi-scale convolution, and performs data enhancement on the structural vibration response information collected by each sensor based on the sliding window overlapping method; echo state network and multi-scale convolution neural network The joint model judges the damage state based on the temporal dependence characteristics of the structural vibration response information collected by each sensor and the spatial correlation characteristic information between the structural vibration response information collected by different sensors. The invention can effectively extract the time-sequence dependence and spatial correlation between the structural vibration response data, thereby accurately judging whether the structure is damaged and the degree of damage. The method is scientific and efficient; it can estimate the damage nature of the structure in real time, and grasp the structural safety status, prevent disasters and accidents, and ensure the safety of structural engineering operations.

Description

technical field [0001] The invention relates to a structural engineering and safety monitoring method, in particular to a structural damage identification method based on an echo state and multi-scale convolution joint model. Background technique [0002] With the rapid development of my country's economy, a large number of bridges, high-rise buildings, ocean platforms, water conservancy facilities, aviation aircraft and other important structural projects have been built in China. These structural projects have the characteristics of complex and changeable service environments and large volumes. During its non-stop operation for decades or even hundreds of years, due to the influence of natural factors such as earthquakes, corrosion of the external environment, and aging of its own materials, or due to human factors such as load fatigue, poor supervision, and design defects, It is easy to cause problems such as local damage and lower bearing capacity of the structure. In s...

AI Technical Summary

Problems solved by technology

Therefore, the problem of how to construct more samples based on the simulation and laboratory real bridge model data needs to be solved urgently; (3) The shallow neural network model has a weak feature extraction ability due to insufficient depth of the model, thus damage identification The accuracy is low; the deep neural network model tends to cause the network gradient to disappear or the calculation to be complex as the depth of the model increases

Images