Traditional physical model-based nondestructive evaluation (NDE) and damage detection methods are often unreliable due to the complex dependence of model parameters on minor differences in material properties (e.g., thickness, temperature, or loading effects). While classic data-driven approaches appear to eliminate model complexity, their performance highly depends on feature extraction, for which domain-expertise-based data preprocessing is required. Wavefield analysis is a promising alternative for non-contact NDE but suffers from the problem of slow data acquisition. As a result, effective structural health monitoring (SHM) based on wavefield analysis of guided waves in large-scale systems, such as mechanical, civil, or aerospace structures, has remained challenging. To address these challenges, we present a deep convolutional neural network (DCNN)-based transfer learning approach to interpret ultrasonic guided waves with small training data sets, thereby achieving rapid, effective, and automated SHM. Specifically, the proposed learning framework includes a pre-trained DCNN for automated feature extraction from the raw inputs (i.e., wavelet-transformed time-frequency images) and a fully connected classification stage that is trained with partial wavefield scans. Experiments on full wavefield scans of various thin metal plates demonstrate the effectiveness and efficiency of the proposed approach: >95% classification accuracy is obtained with only 5% training data, thus enabling fast scanning and fully automated damage detection of large-scale structures.
Structural health monitoring (SHM) enables wide-area, in situ, and continuous evaluation of the health of mechanical, civil, and aerospace structures to detect the existence, location, and severity of damage. In this paper, we introduce a sparse and scalable sensor network, driven by custom ultra-low power and highly integrated CMOS transceivers, that is suitable for a variety of active SHM applications using ultrasonic guided waves. The transceivers both generate electrical actuation signals for piezoelectric transducers and provide broadband lownoise reception of returned signals. Specifically, the transmitter can generate narrow-band Hanning-windowed sinusoids (5 cycles long) up to 12.7 Vpp with a center frequency in the 100 kHz to ∼2.8 MHz range. The waveform is synthesized using filtered pulse-width modulation (PWM), which is integrated with a programmable phasedlocked loop (PLL) to achieve low distortion and high repeatability. The fully-differential low-noise receiver is capable of performing a Hilbert transform on-chip to extract both amplitude and phase information from the received signal. For long term monitoring, we propose a two-step SHM strategy, in which damage is first detected using environmentally compensated data and is then localized and characterized. Furthermore, we discuss two commonly-used SHM damage localization algorithms, namely RAPID and delay-and-sum, in terms of computation, memory, and power consumption for the proposed sparse SHM networks. The proposed approach has been effectively demonstrated both in simulations and in experiments on an SHM test bed.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
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.