Related Experiment Video
Updated: Jan 15, 2026
11:34
Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography
Published on: May 15, 2017
11.5K
High-precision multiple defect detection and localization in composite laminates using integrated piezoelectric
Majid Ghazali1, Morteza Karamooz Mahdiabadi2
1Department of Mechanical Engineering, Tarbiat Modares University, P.O. Box 14115-177, Tehran, Iran.
Scientific Reports
|October 13, 2025
Summary
This study introduces a new method using piezoelectric sensors and AI to precisely find multiple cracks and delaminations in composite materials. This advanced structural health monitoring system achieves over 99% accuracy in locating defects.
Area of Science:
- Materials Science
- Mechanical Engineering
- Artificial Intelligence
Background:
- Composite laminates are susceptible to delamination and crack defects.
- Effective non-destructive evaluation (NDE) is crucial for structural integrity.
- Existing methods may lack accuracy in detecting multiple, complex defects.
Purpose of the Study:
- To develop a novel methodology for high-accuracy detection and localization of multiple defects in composite laminates.
- To integrate piezoelectric actuation/sensing with advanced computational models.
- To assess the performance of regression and neural network techniques for defect characterization.
Main Methods:
- Utilized an eight-layer graphite/epoxy composite plate instrumented with piezoelectric patches.
- Employed random voltage stimuli for excitation and captured structural responses.
- Integrated six regression techniques and artificial neural networks (ANNs) for localization.
- Combined five signal decomposition methods with four classifiers for defect type identification.
Main Results:
- Achieved localization accuracy exceeding 99.6% using the proposed framework.
- Attained up to 98.26% accuracy in identifying defect types.
- Demonstrated precise detection of previously unseen delamination and crack defects at the lamina level.
- Found piezoelectric sensor voltage signals superior to acceleration signals for defect characterization.
Conclusions:
- The integrated piezoelectric-regression/neural network framework provides robust non-destructive evaluation capabilities.
- The methodology shows significant potential for real-time structural health monitoring.
- This approach offers a precise and reliable solution for composite material integrity assessment.
