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 se
- PDF / 662,421 Bytes
- 6 Pages / 595.516 x 790.987 pts Page_size
- 76 Downloads / 191 Views
Experimental Study on Digital Image Correlation for Deep Learning-Based Damage Diagnostic Nur Sila Gulgec, Martin Takáˇc, and Shamim N. Pakzad
Abstract 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. Keywords Structural health monitoring · Digital image correlation · Convolutional neural networks · Damage detection
28.1 Introduction It is important to establish lifetime safety of the infrastructure subjected to a wide range of environmental and operational conditions [1]. Providing timely damage assessment of these structures often requires long-term monitoring and dense instrumentation [2]. Sensor networks today provide an exciting set of opportunities and challenges to collect an enormous amount of data from any structure, which due to its nature is posing a big data problem [3]. Conventional approaches primarily focus on hand-crafting damage features and classifiers to interpret the health condition of the structures [4–7]. Although such methods are effective in identifying structural damage of a particular type, there are some constraints limiting these methods. The existing methods of analysis rely on estimating carefully crafted features that often are limited in what they can do and are not automated in nature, thus not appropriate for a broad range of big data applications [8, 9]. Deep Neural Networks (deep learning or DNN) are a state-of-the-art set of methods for taking advantage of the opportunities hidden in big data. They are designed such that they can learn from data, for this reason, deep learning is ideally suited to use large representative training datasets to learn complex features [10]. DNNs learn by training the network parameters by training which is then used to make data-driven predictions or decisions. One of the most widely used types of DNN is convolutional neural network (CNN) due to its ability to keep spatial features of the input and reduce memory requirements by using fewer par
Data Loading...
