.png){kind=icon}
Return
Damage localization and quantification in plate structures using ensemble network
.png){kind=icon}
OA
AI
DOI：10.1016/j.engstruct.2024.119146.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The development of structural health monitoring has ramped up significantly in the era of machine learning. A seemingly healthy structure can possess imperceptible internal defects of any kind. Therefore, a quick and reliable damage detection system is of interest to determine if maintenance action is needed. In this study, an ensemble network involving a Convolutional Neural Network (CNN) modified from GoogLeNet is tested on two plate structures, one isotropic and the other orthotropic. The isotropic plate damage localization is readily obtained from a modified regression network, while the quantification is obtained from the ensemble network for improved accuracy. The orthotropic case also identifies the layer in which the damage occurs through an extra output parameter from the model, which makes it necessary to build a separate localization network that is assembled using the ensemble network for general cases. The accuracy of the results obtained using the ensemble network is far better than those based on the use of individual network altogether. Overall, the localization and quantification processes are accurately completed for both plate structures.
Keywords:
CNN
Damage detection
Ensemble network
Laminate
Thin Plate
Vibration-based analysis
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
6.4
Papers:
2.1W
Citations:
8.7W
Organization
No organization information available
Cited Papers
Structural damage detection based on convolutional neural networks and population of bridges.png)
MEASUREMENT
MEASUREMENTIF5.6
PREAI
ShareSaveShareSaveAutomatic void content assessment of composite laminates using a machine-learning approachCOMPOSITE STRUCTURESOAAI
COMPOSITE STRUCTURESIF7.1
OAAIShareSaveEfficient Artificial neural networks based on a hybrid metaheuristic optimization algorithm for damage detection in laminated composite structuresCOMPOSITE STRUCTURES
COMPOSITE STRUCTURESIF7.1
PREAI
ShareSaveFault diagnosis on beam-like structures from modal parameters using artificial neural networksMEASUREMENT
MEASUREMENTIF5.6
PREAI
ShareSaveVibration-based inverse algorithms for detection of delamination in compositesCOMPOSITE STRUCTURES
COMPOSITE STRUCTURESIF7.1
PREAI
ShareSave
A modified transmissibility indicator and Artificial Neural Network for damage identification and quantification in laminated composite structuresCOMPOSITE STRUCTURES
COMPOSITE STRUCTURESIF7.1
PREAI
ShareSave
View more