scholarly journals A Novel ANN-RDT Approach for Damage Detection of a Composite Panel Employing Contact and Non-Contact Measuring Data

2021 ◽  
pp. 114794
Author(s):  
Alireza Mojtahedi ◽  
Hamid Hokmabady ◽  
Mohsen Kouhi ◽  
Samira Mohammadyzadeh
2019 ◽  
Vol 19 (5) ◽  
pp. 1524-1541 ◽  
Author(s):  
Alessandro Marzani ◽  
Nicola Testoni ◽  
Luca De Marchi ◽  
Marco Messina ◽  
Ernesto Monaco ◽  
...  

This article reports on the creation of an open database of piezo-actuated and piezo-received guided wave signals propagating in a composite panel of a full-scale aeronautical structure. The composite panel closes the bottom part of a wingbox that, along with the leading edge, the trailing edge, and the wingtip, forms an outer wing demonstrator approximately 4.5 m long and from 1.2 to 2.3 m wide. To create the database, a structural health monitoring system, composed of a software/hardware central unit capable of controlling a network of 160 piezoelectric transducers secondarily bonded on the composite panel, has been realized. The structural health monitoring system has been designed to (1) perform electromechanical impedance measurement at each transducer, in order to check for their reliability and bonding strength, and (2) to operate an active guided wave screening for damage detection in the composite panel. Electromechanical impedance and guided wave measurements were performed at four different testing stages: before loading, before fatigue, before impacts, and after impacts. The database, freely available at http://shm.ing.unibo.it/ , can thus be used to benchmarking, on real-scale structural data, guided wave algorithms for loading, fatigue, as well as damage detection, characterization, and sizing. As an example, in this work, a delay and sum algorithm is applied on the post-impact data to illustrate how the database can be exploited.


2016 ◽  
Vol 713 ◽  
pp. 244-247 ◽  
Author(s):  
Hidekazu Tanaka ◽  
Zahra Sharif Khodaei

Probability-based imaging which illustrates a distribution map of probability of damage presence in structures is a diagnostic method well established for damage detection in sensorized structures. Since the quality of the recorded signal is directly linked to the reliability of the diagnostic outcome, the assessment of robustness of the damage detection methodology is of high significance. In this paper, robustness and reliability of the current probability based imaging algorithms have been assessed for detecting BVID in a composite panel. Consequently, a proposed outlier analysis and DI probability distribution damage detection algorithm was shown to improve the reliability of the detection method.


2012 ◽  
Vol 518 ◽  
pp. 160-167 ◽  
Author(s):  
Andrzej Klepka ◽  
Wieslaw Jerzy Staszewski ◽  
T. Uhl ◽  
Dario di Maio ◽  
Fabrizio Scarpa ◽  
...  

This paper demonstrates impact damage detection in a composite sandwich panel. The panel is built from a chiral honeycomb and two composite skins. Chiral structures are a subset of auxetic solids exhibiting counterintuitive deformation mechanism and rotative but not reflective symmetry. Damage detection is performed using nonlinear acoustics,involves combined vibro-acoustic interaction of high-frequency ultrasonic wave and low-frequency vibration excitation. High-and low-frequency excitations are introduced to the panel using a low-profile piezoceramic transducer and an electromagnetic shaker, respectively. Vibro-acoustic modulated responses are measured using laser vibrometry. The methods used for impact damage detection clearly reveal de-bonding in the composite panel. The high-frequency weak ultrasonic wave is also modulated by the low-frequency strong vibration wave when nonlinear acoustics is used for damage detection. As a result frequency sidebands can be observed around the main acoustic harmonic in the spectrum of the ultrasonic signal.


2016 ◽  
Vol 713 ◽  
pp. 191-194 ◽  
Author(s):  
Laure Sainfort ◽  
Zahra Sharif Khodaei ◽  
M.H. Ferri Aliabadi

In this work the optimal configuration of transducers for damage detection and localization has been investigated. A particular interest is given to three optimization methods: mini-max, average Probability of Non Detection (POND) and ray tracing approach, coupled with genetic algorithm. After optimal configurations have been computed for each technique, they are experimentally tested and compared on a composite panel with one or two damages by generating and receiving Lamb waves signals. Damage detection is carried out with the Probability Based Damage Index Method (PBDIM). It was found that, in most cases, the ray tracing method and the average POND technique give better results, with a good detection of damages in comparison to the minimax POND technique, even if the latter seems numerically better.


Sign in / Sign up

Export Citation Format

Share Document