Damage identification method and equipment based on two-dimensional convolutional neural network

A damage identification and neural network technology, applied in the field of damage identification based on two-dimensional convolutional neural network, can solve the problems of staying at the qualitative level, difficult to achieve quantitative analysis and characterization of structural damage, and achieve high accuracy and feature extraction capabilities. Strong, improve the effect of prediction accuracy

Pending Publication Date: 2022-03-15
HUAZHONG UNIV OF SCI & TECH
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] However, piezoelectric impedance technology also has its problems. Although this method can use the changes in the impedance spectrum measured by piezoelectric sheets to reflect structural damage and realize damage identification of complex structures, the judgment of damage is mostly at the qualitative level, which is difficult to achieve. Quantification and characterization of structural damage

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Damage identification method and equipment based on two-dimensional convolutional neural network
  • Damage identification method and equipment based on two-dimensional convolutional neural network
  • Damage identification method and equipment based on two-dimensional convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0054] A damage recognition method based on two-dimensional convolutional neural network, such as figure 2 shown, including:

[0055] Sweep in the preset frequency band and measure the conductance signal of the structure at each frequency point;

[0056] Divide the preset frequency band into a plurality of sub-frequency bands, calculate the deviation of the conductance signal of the structure in each sub-frequency band relative to the conductance signal in each sub-frequency band of the structure in an undamaged state, and use the n sub-frequency bands with the largest deviation as the target frequency band; Each frequency band contains N frequency points, both n and N are positive integers, and n≥2;

[0057] Extract the conductance signals with structures in n target frequency bands, construct the one-dimensional conductance signals in every two target frequency bands into N×N two-dimensional data, and form a prediction sample from the constructed two-dimensional data; in t...

Embodiment 2

[0090] A damage recognition device based on a two-dimensional convolutional neural network, including: a health monitoring module, a frequency band screening module, a data processing module, and a damage recognition module;

[0091] The health monitoring module is used to sweep the frequency in the preset frequency band and measure the conductance signal of the structure at each frequency point;

[0092] The frequency band screening module is used to divide the preset frequency band into a plurality of sub-frequency bands, calculate the deviation of the conductance signal of the structure in each sub-frequency band relative to the conductance signal of the structure in each sub-frequency band in an undamaged state, and select the n sub-bands with the largest deviation The frequency band is used as the target frequency band; each sub-frequency band contains N frequency points, both n and N are positive integers, and n≥2;

[0093] The data processing module is used for extracti...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a damage identification method and device based on a two-dimensional convolutional neural network, and belongs to the field of structural health monitoring, and the method comprises the steps: carrying out the frequency sweeping in a preset frequency band, and measuring the conductance signal of a structure at each frequency point; dividing the preset frequency band into a plurality of sub-frequency bands, calculating the deviation of the conductance signal in each sub-frequency band relative to the conductance signal in each sub-frequency band in a non-damage state, and taking the n sub-frequency bands with the maximum deviation as a target frequency band; extracting one-dimensional conductance signals under n target frequency bands, and constructing two-dimensional data by every two one-dimensional conductance signals to obtain a prediction sample; inputting the prediction sample into a trained damage identification model so as to predict the current quality loss amount of the structure and obtain the damage state of the structure; wherein the damage identification model is a two-dimensional convolutional neural network and is used for predicting the mass loss amount of the structure through two-dimensional data. According to the method, the different electrical impedance information can be accurately classified and quantified, and quantitative analysis and characterization of structural damage are realized.

Description

technical field [0001] The invention belongs to the field of structural health monitoring, and more specifically relates to a damage identification method and equipment based on a two-dimensional convolutional neural network. Background technique [0002] With the rapid development of infrastructure construction, various large-scale civil and industrial buildings have emerged one after another, and gradually move towards diversification, sophistication, complexity and internationalization. At the same time as the development of my country's architecture, people's requirements for structural safety have increased with it. For all kinds of buildings at present, the loads are gradually complicated, including direct loads, indirect loads and accidental loads, and the coupling of this series of loads often leads to structural fatigue damage, structural damage and resistance decline. In addition, the building may also be affected by its own and external factors during its use, su...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F2218/16G06F2218/02G06F18/2414G06F18/214
Inventor 艾德米莫芳程佳宝朱宏平
Owner HUAZHONG UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products