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.
The impedance/admittance measurements of a piezoelectric transducer circuit bonded to or embedded in a host structure can be used as damage indicator, since damage will introduce notable impedance shifts. When a credible model of the healthy structure, such as the finite element model, is available, using the impedance/admittance change information as input, it is possible to identify both the location and severity of damage. In this research we cast the damage identification problem into a many-objective optimization framework through impedance response calibration using Gaussian Process. With damage location and severity as unknown variables, the objective functions are response surfaces calibrated using emulated damaged scenarios assisted by Gaussian Process. Subsequently, a ε - dominance enabled many-objective algorithm based on multi-objective Simulated Annealing is devised to facilitate the many-objective optimization. The proposed approach yields high-quality results that can be further investigated for model updating.
Structural damage identification has been continuously pursued in engineering practices to facilitate diagnosis and prognosis in structural health monitoring (SHM) systems. In SHM, the changes of modal parameters are frequently used as inputs. In this research, we employ the multiple damage location assurance criterion (MDLAC) to characterize the correlation between predictions of both frequency changes and single mode shape change with the measured data. The damage locations and severities can be obtained by maximizing the MDLAC values. Thereafter, a multi-objective optimization problem based on their MDLAC values can be formulated and optimized by applying a newly devised multi-objective DIRECT approach. The proposed approach offers practical attractions of only requiring a short amount of computational time, and the results are conclusive and repeatable.
For large-scale wind turbines, reducing maintenance cost is a major challenge. Model predictive control (MPC) is a promising approach to deal with multiple conflicting objectives using the weighed sum approach. In this research, model predictive control method is applied to wind turbine to find an optimal balance between multiple objectives, such as the energy capture, loads on turbine components, and the pitch actuator usage. The actuator constraints are integrated into the objective function at the control design stage. The analysis is carried out in both the partial load region and full load region, and the performances are compared with those of a baseline gain scheduling PID controller. The application of this strategy achieves enhanced balance of component loads, the average power and actuator usages in partial load region.
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.