This paper proposes a deep learning-based rapid inspection method for concrete structures. The proposed method is composed of three steps: (1) collection of a large volume of images containing damage information from internet, (2) development of a deep learning model (i.e., convolutional neural network (CNN)) using collected images, and (3) automatic selection of damage images using the trained deep learning model. In the first step, the internet-based search benefits in easy classification of collected images by varying searching word, and in collection of images taken under diverse environmental conditions. In the second step, a transfer learning approach has been introduced to save the time and cost for developing a deep learning model. In the third step, the probability map is introduced based on duplicated searching to make the searching process robust. The whole procedure of the proposed method has been applied to some figures taken in a real structure. This method is expected to facilitate the regular inspection and speed up the assessment of detailed damage distribution the without losing accuracy.
This paper presents a deep learning-based concrete crack detection technique using hybrid images. The hybrid images combining vision and infrared (IR) thermography images are able to improve crack detectability while minimizing false alarms. Large scale concrete-made infrastructures such as bridge, dam, and etc. can be effectively inspected by spatially scanning the hybrid imaging system including vision camera, IR camera and continuous-wave line laser. However, the decision-making for the crack identification often requires experts’ intervention. As a target concrete structure gets larger, automated decision-making becomes more necessary in the practical point of view. The proposed technique is able to achieve automated crack identification by modifying a well-trained convolutional neural network using a set of crack images as a training image set, while retaining the advantages of hybrid images. The proposed technique is experimentally validated using a lab-scale concrete specimen developed with various-size cracks. The test results reveal that macro- and micro-cracks are automatically detected with minimizing false-alarms.
As the number of long-span bridges is increasing worldwide, maintaining their structural integrity and safety become an important issue. Because the stay cable is a critical member in most long-span bridges and vulnerable to wind-induced vibrations, vibration mitigation has been of interest both in academia and practice. While active and semi-active control schemes are known to be quite effective in vibration reduction compared to the passive control, requirements for equipment including data acquisition, control devices, and power supply prevent a widespread adoption in real-world applications. This study develops an integrated system for vibration control of stay-cables using wireless sensors implementing a semi-active control. Arduino, a low-cost single board system, is employed with a MEMS digital accelerometer and a Zigbee wireless communication module to build the wireless sensor. The magneto-rheological (MR) damper is selected as a damping device, controlled by an optimal control algorithm implemented on the Arduino sensing system. The developed integrated system is tested in a laboratory environment using a cable to demonstrate the effectiveness of the proposed system on vibration reduction. The proposed system is shown to reduce the vibration of stay-cables with low operating power effectively.
System identification is a fundamental process for developing a numerical model of a physical structure. The system
identification process typically involves in data acquisition; particularly in civil engineering applications accelerometers
are preferred due to its cost-effectiveness, low noise, and installation convenience. Because the measured acceleration
responses result in translational degrees of freedom (DOF) in the numerical model, moment-resisting structures such as
beam and plate are not appropriately represented by the models. This study suggests a system identification process that
considers both translational and rotational DOFs by using accelerometers and gyroscopes. The proposed approach
suggests a systematic way of obtaining dynamic characteristics as well as flexibility matrix from two different
measurements of acceleration and angular velocity. Numerical simulation and laboratory experiment are conducted to
validate the efficacy of the proposed system identification process.
KEYWORDS: Bridges, Sensors, Smart sensors, Sensor networks, Structural health monitoring, Antennas, Data processing, Solar cells, Connectors, Data transmission
Cables are critical load carrying members of cable-stayed bridges; monitoring tension forces of the cables provides valuable information for SHM of the cable-stayed bridges. Monitoring systems for the cable tension can be efficiently realized using wireless smart sensors in conjunction with vibration-based cable tension estimation approaches. This study develops an automated cable tension monitoring system using MEMSIC’s Imote2 smart sensors. An embedded data processing strategy is implemented on the Imote2-based wireless sensor network to calculate cable tensions using a vibration-based method, significantly reducing the wireless data transmission and associated power consumption. The autonomous operation of the monitoring system is achieved by AutoMonitor, a high-level coordinator application provided by the Illinois SHM Project Services Toolsuite. The monitoring system also features power harvesting enabled by solar panels attached to each sensor node and AutoMonitor for charging control. The proposed wireless system has been deployed on the Jindo Bridge, a cable-stayed bridge located in South Korea. Tension forces are autonomously monitored for 12 cables in the east, land side of the bridge, proving the validity and potential of the presented tension monitoring system for real-world applications.
Rapid advancement of sensor technology has been changing the paradigm of Structural Health Monitoring (SHM)
toward a wireless smart sensor network (WSSN). While smart sensors have the potential to be a breakthrough to current
SHM research and practice, the smart sensors also have several important issues to be resolved that may include robust
power supply, stable communication, sensing capability, and in-network data processing algorithms. This study is a
hybrid WSSN that addresses those issues to realize a full-scale SHM system for civil infrastructure monitoring. The
developed hybrid WSSN is deployed on the Jindo Bridge, a cable-stayed bridge located in South Korea as a continued
effort from the previous year's deployment. Unique features of the new deployment encompass: (1) the world's largest
WSSN for SHM to date, (2) power harvesting enabled for all sensor nodes, (3) an improved sensing application that
provides reliable data acquisition with optimized power consumption, (4) decentralized data aggregation that makes the
WSSN scalable to a large, densely deployed sensor network, (5) decentralized cable tension monitoring specially
designed for cable-stayed bridges, (6) environmental monitoring. The WSSN implementing all these features are
experimentally verified through a long-term monitoring of the Jindo Bridge.
KEYWORDS: Bridges, Smart sensors, Structural health monitoring, Sensors, Sensor networks, Wind measurement, Data communications, Energy harvesting, Solar cells, Head
This paper presents a structural health monitoring (SHM) system using a dense array of scalable smart wireless sensor
network on a cable-stayed bridge (Jindo Bridge) in Korea. The hardware and software for the SHM system and its
components are developed for low-cost, efficient, and autonomous monitoring of the bridge. 70 sensors and two base
station computers have been deployed to monitor the bridge using an autonomous SHM application with consideration of
harsh outdoor surroundings. The performance of the system has been evaluated in terms of hardware durability, software
reliability, and power consumption. 3-D modal properties were extracted from the measured 3-axis vibration data using
output-only modal identification methods. Tension forces of 4 different lengths of stay-cables were derived from the
ambient vibration data on the cables. For the integrity assessment of the structure, multi-scale subspace system
identification method is now under development using a neural network technique based on the local mode shapes and
the cable tensions.
For the deflection measurement of a bridge, sensor inaccessibility due to the reference point (fixed point) may cause lots
of inconvenience and increase of installation cost and time. Therefore, an alternative indirect deflection estimation
method is proposed utilizing the FE model updating with measured acceleration data from an ambient vibration test. For
the fast and stable optimization during the model updating, two step model updating approach is additionally proposed.
To investigate the effects of used modal quantities and their weighting factors, extensive model updating was carried out
by applying several kinds of data sets: using only natural frequencies, using only mode shapes, and using both of them,
By simulating the load test on each of the updated model, deflections were estimated and compared with the real
measured deflection data in the load test.
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.