Damage diagnosis has been a challenging inverse problem in structural health monitoring. The main difficulty is characterizing the unknown relation between the measurements and damage patterns (i.e., damage indicator selection). Such damage indicators would ideally be able to identify the existence, location, and severity of damage. Therefore, this procedure requires complex data processing algorithms and dense sensor arrays, which brings computational intensity with it. To address this limitation, this paper introduces convolutional neural network (CNN), which is one of the major breakthroughs in image recognition, to the damage detection and localization problem. The CNN technique has the ability to discover abstract features and complex classifier boundaries that are able to distinguish various attributes of the problem. In this paper, a CNN topology was designed to classify simulated damaged and healthy cases and localize the damage when it exists. The performance of the proposed technique was evaluated through the finite-element simulations of undamaged and damaged structural connections. Samples were trained by using strain distributions as a consequence of various loads with several different crack scenarios. Completely new damage setups were introduced to the model during the testing process. Based on the findings of the proposed study, the damage diagnosis and localization were achieved with high accuracy, robustness, and computational efficiency.
more »
« less
Experimental Study on Digital Image Correlation for Deep Learning-Based Damage Diagnostic
Large quantities of data which contain detailed condition information over an extended period of time should be utilized to prioritize infrastructure repairs. As the temporal and spatial resolution of monitoring data drastically increase by advances in sensing technology, structural health monitoring applications reach the thresholds of big data. Deep neural networks are ideally suited to use large representative training datasets to learn complex damage features. In the previous study of authors, a real-time deep learning platform was developed to solve damage detection and localization challenge. The network was trained by using simulated structural connection mimicking the real test object with a variety of loading cases, damage scenarios, and measurement noise levels for successful and robust diagnosis of damage. In this study, the proposed damage diagnosis platform is validated by using temporally and spatially dense data collected by Digital Image Correlation (DIC) from the specimen. Laboratory testing of the specimen with induced damage condition is performed to evaluate the performance and efficiency of damage detection and localization approach.
more »
« less
- Award ID(s):
- 1618717
- NSF-PAR ID:
- 10110701
- Date Published:
- Journal Name:
- Proceedings of the Society for Experimental Mechanics
- Volume:
- 1
- ISSN:
- 1046-6789
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract. In this paper, two different convolutional neural networks (CNNs) are applied on images for automated structural damage detection (SDD) in earthquake damaged structures and cracking localization (e.g., detection of cracks, their widths and distributions) at various scales, such as pixel level, object level, and structural level. The proposed method has two main steps: 1) diagnosis, and 2) localization of cracking or other damage. At first a residual CNN with transfer learning is employed to classify the damage in the structures and structural components. This step performs damage detection using two public datasets. The second step uses another CNN with U-Net structure to locate the cracking on low resolution images. The implementations using public and self-collected datasets show promising performance for a problem that had remained a challenge in the structure engineering field for a long time and indicate that the proposed approach can perform detection and localization of structural damage with an acceptable accuracy.more » « less
-
Damage detection and localization remain challenging research areas in structural health monitoring. Guided wave-based methods that utilize signal processing tools (e.g., matched field processing and delay-and-sum localization) have enjoyed success in damage detection. To locate damage, such techniques rely on a model of wave propagation through materials. Measured data is then compared with these models to determine the origin of a wave. As a result, the analytical model and actual data may have a mismatch due to environmental variations or a lack of knowledge about the material. Deep neural networks are a class of machine learning algorithms that learn a non-linear functional mapping. The paper presents a deep neural network-based approach to damage localization. We use simulated data to assess the performance of localization frameworks under varying levels of noise and other uncertainty in our ultrasonic signals.more » « less
-
more » « less
Recent advances in energy harvesting technologies have led to the evolution of self-powered structural health monitoring techniques that are energy-efficient. Concurrent to the emergence of self-powered sensing has been the development of power-efficient data communication protocols. One such technology is an energy-aware pulse communication architecture that employs ultrasonic pulses through the material substrate for information forwarding. This results in limited discrete binary data that raises the need for new data analysis methods for structural health monitoring purposes. A pattern recognition framework that allows interpretation of the resulting asynchronous discrete binary data for condition and damage assessment in plate-like structures is presented in this article. The proposed pattern recognition framework is based on integration of image-based pattern recognition using anomaly detection, a pattern anomaly measure, a focal density concept, and the k-nearest neighbor algorithm. Using numerical simulations, damage indication parameters were determined from the strain response of dynamically loaded plates. Simulated test cases considering different levels of damage severity, single and multiple damage regions, loading conditions, and measurement noise were studied to evaluate the effectiveness and robustness of the strategy. Furthermore, the effect of sensor density on the proposed strategy was explored. Results demonstrate satisfactory performance and robustness of the proposed pattern recognition framework for localized damage detection in plate-like structures using limited and low-resolution discrete binary data.
-
Vision-based structural health monitoring (SHM) has become an important approach to recognize and evaluate structural damage after natural disasters. Deep convolutional neural networks (CNNs) have recently attained a breakthrough in computer vision field, in particular for image classification task. In this article, we adopted deep residual neural network (ResNet) whose residual representations and shortcut connections mechanism has gained significant performance in various computer vision tasks. In addition, we applied transfer learning due to a relatively small number of training images. To test our approach, we used the dataset from the 2018 PEER Hub ImageNet Challenge distributed by Pacific Earthquake Engineering Research Center. This challenge proposed eight structural damage detection tasks: scene classification, damage check, spalling condition, material type, collapse check, component type, damage level and damage type which can be categorized as binary and multi-class (3 or 4 classes) classification problems. Our experiments with eight different tasks showed that reliable classification can be obtained for some tasks. Corresponding above eight tasks, classification accuracy varied from 63.1% to 99.4%. Our approach has attained third place for overall tasks in this challenge. Through the individual observation of training dataset, it is found that there are a large number of confusing images. Therefore, it is believed that the accuracy will be improved after making a precise training data.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1007/978-3-030-12115-0_28
