Journal
MEASUREMENT
Volume 200, Issue -, Pages -Publisher
ELSEVIER SCI LTD
DOI: 10.1016/j.measurement.2022.111680
Keywords
Ultrasonic guided waves; Elastic constants; Deep learning; Dispersion curves; Composites
Ask authors/readers for more resources
In this paper, a deep learning model called Elasticity Network (ENet) is proposed to rapidly characterize the elastic constants of composite laminates. The superiority and robustness of the proposed method are demonstrated through numerical and experimental verifications.
An immediate and convenient report of mechanical properties of composites with full automation is crucial for timely characterizing the time-dependent degradation of material properties and assessing structural fatigues. This paper aims to rapidly characterize the elastic constants of composite laminates with ultrasonic guided waves. We propose a deep learning model namely Elasticity Network (ENet) to characterize composites by correlating its elastic constants directly with guided wave dispersion curves. With two adjacent guided wave signals, continuous reconstructed dispersion curve segments are fed into the well-trained network to output the real-time display of mechanical properties. ENet is trained with theoretical data, thereby allowing the extensive applications of the proposed method without the need of tedious data collection. It also eliminates the common necessity for multi-directional guided wave measurements to characterize anisotropic properties. Numerical and experimental verifications are conducted on fiber-reinforced composite laminates to demonstrate the superiority and robustness of the proposed method.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available