Bridge-vehicle coupled vibration response and static test data based damage identification of highway bridges

2013 ◽  
Vol 46 (1) ◽  
pp. 75-90 ◽  
Author(s):  
Jinsong Zhu ◽  
Qiang Yi
Keyword(s):  
Test Data ◽  
Damage Identification ◽  
Highway Bridges ◽  
Vibration Response ◽  
Coupled Vibration ◽  
Static Test

Damage identification from static tests by eigenparameter decomposition and sparse regularization

2019 ◽  
Vol 19 (5) ◽  
pp. 1351-1374
Author(s):  
Zhong-Rong Lu ◽  
Junxian Zhou ◽  
Li Wang ◽  
Jike Liu
Keyword(s):  
Test Data ◽  
Damage Identification ◽  
Regularization Parameter ◽  
Goal Function ◽  
Structural Safety ◽  
Sparse Regularization ◽  
Static Tests ◽  
Static Test ◽  
Static Data ◽  
Identification Approach

Identifying the damages from test data is central to assuring the structural safety. The static model is the simplest model to describe the mechanical behavior of the structure where only the stiffness is involved and it is independent of the mass and the complex damping. As a result, damage identification based on the static data will not be deteriorated by the inexact damping and the possible error in the mass. Notwithstanding, the major difficulty regarding damage identification with static test data is that the amount of the static data is quite limited and insufficient with respect to the amount of damage parameters, rendering the identification very sensitive to the measurement noise. Attempting to circumvent this difficulty, a novel damage identification approach is developed in this article where the sparse regularization is introduced to implicitly enforce the sparsity constraint of the damage locations. Moreover, in order to work well with the sparse regularization, a new goal function is established by resorting to the eigenparameter decomposition for which the decoupling feature would make the sparse regularization be tackled immediately with closed-form solutions. Then, the alternating minimization approach is used to get the solution of the new goal function and the threshold setting method is simply called to determine a proper regularization parameter. Numerical and experimental examples are studied to testify the feasibility, accuracy, and robustness of the proposed damage identification approach.

scholarly journals Structural Damage Identification of Composite Rotors Based on Fully Connected Neural Networks and Convolutional Neural Networks

Sensors ◽  
2021 ◽  
Vol 21 (6) ◽  
pp. 2005
Author(s):  
Veronika Scholz ◽  
Peter Winkler ◽  
Andreas Hornig ◽  
Maik Gude ◽  
Angelos Filippatos
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Composite Structures ◽  
Damage Identification ◽  
Vibration Response ◽  
Damage Initiation ◽  
Matrix Cracks ◽  
Operational Life ◽  
Multiple Data Sets ◽  
Fully Connected

Damage identification of composite structures is a major ongoing challenge for a secure operational life-cycle due to the complex, gradual damage behaviour of composite materials. Especially for composite rotors in aero-engines and wind-turbines, a cost-intensive maintenance service has to be performed in order to avoid critical failure. A major advantage of composite structures is that they are able to safely operate after damage initiation and under ongoing damage propagation. Therefore, a robust, efficient diagnostic damage identification method would allow monitoring the damage process with intervention occurring only when necessary. This study investigates the structural vibration response of composite rotors by applying machine learning methods and the ability to identify, localise and quantify the present damage. To this end, multiple fully connected neural networks and convolutional neural networks were trained on vibration response spectra from damaged composite rotors with barely visible damage, mostly matrix cracks and local delaminations using dimensionality reduction and data augmentation. A databank containing 720 simulated test cases with different damage states is used as a basis for the generation of multiple data sets. The trained models are tested using k-fold cross validation and they are evaluated based on the sensitivity, specificity and accuracy. Convolutional neural networks perform slightly better providing a performance accuracy of up to 99.3% for the damage localisation and quantification.

Sign in / Sign up

Export Citation Format

Share Document