Guided waves-based SHM using an ML-based parametric ROM

Paul Sieber; Konstantinos Agathos; Rohan Soman; Wieslaw Ostachowicz; Eleni Chatzi

doi:10.1117/12.2658304

18 April 2023 Guided waves-based SHM using an ML-based parametric ROM

Paul Sieber, Konstantinos Agathos, Rohan Soman, Wieslaw Ostachowicz, Eleni Chatzi

Proceedings Volume 12487, Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XVII; 1248704 (2023) https://doi.org/10.1117/12.2658304
Event: SPIE Smart Structures + Nondestructive Evaluation, 2023, Long Beach, California, United States

Abstract

Structural Health Monitoring (SHM) using Ultrasonic Guided Waves (UGWs) offers great potential in detection of minor flaws, due to the employed short wavelengths. A bottleneck in UGWs-based schemes lies in the extensive computational costs for evaluating the associated wave propagation models. Such detailed models form though a necessity to reach higher levels of SHM, e.g. localization and assessment of flaws. Reduced Order Models (ROMs) and surrogate models allow for lowering the substantial numerical costs for SHM applications, especially if they are parameterized with respect to the characteristics of different flaw configurations. Machine Learning (ML) algorithms can be trained for this purpose, however, in the case of black box ML algorithms, this comes with the drawback of the requirement for substantial data availability for the purpose of training. Such, training data, which are typically derived from full order numerical simulations, are computationally costly to obtain. To reduce the amount of training data, known information on the mechanical behavior can be harnessed and inserted into the estimation process. In the present work, a method is introduced that exploits the properties of the interaction of UGWs with flaws in the frequency domain. It can be shown that the frequency domain response is characterized by periodic features that are linked to the flaw location. An ML model based on this knowledge can be trained with less training data. The potential of this approach for damage localization in the context of SHM is illustrated in a simulated example of a composite plate.

Citation Download Citation

Paul Sieber, Konstantinos Agathos, Rohan Soman, Wieslaw Ostachowicz, and Eleni Chatzi "Guided waves-based SHM using an ML-based parametric ROM", Proc. SPIE 12487, Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XVII, 1248704 (18 April 2023); https://doi.org/10.1117/12.2658304

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
7 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Education and training

Sensors

Structural health monitoring

General packet radio service

Data modeling

Frequency response

Actuators

Show All Keywords

Keywords/Phrases

Search In:

Publication Years