Proc. SPIE. 10168, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2017
KEYWORDS: Information fusion, Safety, Roads, Data modeling, Sensors, Databases, Computing systems, Clouds, Sensor networks, Structural health monitoring, Finite element methods, Bridges, System integration, Web services
This paper describes a cloud-based cyberinfrastructure framework for the management of the diverse data involved in bridge monitoring. Bridge monitoring involves various hardware systems, software tools and laborious activities that include, for examples, a structural health monitoring (SHM), sensor network, engineering analysis programs and visual inspection. Very often, these monitoring systems, tools and activities are not coordinated, and the collected information are not shared. A well-designed integrated data management framework can support the effective use of the data and, thereby, enhance bridge management and maintenance operations. The cloud-based cyberinfrastructure framework presented herein is designed to manage not only sensor measurement data acquired from the SHM system, but also other relevant information, such as bridge engineering model and traffic videos, in an integrated manner. For the scalability and flexibility, cloud computing services and distributed database systems are employed. The information stored can be accessed through standard web interfaces. For demonstration, the cyberinfrastructure system is implemented for the monitoring of the bridges located along the I-275 Corridor in the state of Michigan.
This paper describes an information repository to support bridge monitoring applications on a cloud computing platform.
Bridge monitoring, with instrumentation of sensors in particular, collects significant amount of data. In addition to
sensor data, a wide variety of information such as bridge geometry, analysis model and sensor description need to be
stored. Data management plays an important role to facilitate data utilization and data sharing. While bridge information
modeling (BrIM) technologies and standards have been proposed and they provide a means to enable integration and
facilitate interoperability, current BrIM standards support mostly the information about bridge geometry. In this study,
we extend the BrIM schema to include analysis models and sensor information. Specifically, using the OpenBrIM
standards as the base, we draw on CSI Bridge, a commercial software widely used for bridge analysis and design, and
SensorML, a standard schema for sensor definition, to define the data entities necessary for bridge monitoring
applications. NoSQL database systems are employed for data repository. Cloud service infrastructure is deployed to
enhance scalability, flexibility and accessibility of the data management system. The data model and systems are tested
using the bridge model and the sensor data collected at the Telegraph Road Bridge, Monroe, Michigan.
This paper discusses a data management infrastructure framework for bridge monitoring applications. As sensor technologies mature and become economically affordable, their deployment for bridge monitoring will continue to grow. Data management becomes a critical issue not only for storing the sensor data but also for integrating with the bridge model to support other functions, such as management, maintenance and inspection. The focus of this study is on the effective data management of bridge information and sensor data, which is crucial to structural health monitoring and life cycle management of bridge structures. We review the state-of-the-art of bridge information modeling and sensor data management, and propose a data management framework for bridge monitoring based on NoSQL database technologies that have been shown useful in handling high volume, time-series data and to flexibly deal with unstructured data schema. Specifically, Apache Cassandra and Mongo DB are deployed for the prototype implementation of the framework. This paper describes the database design for an XML-based Bridge Information Modeling (BrIM) schema, and the representation of sensor data using Sensor Model Language (SensorML). The proposed prototype data management framework is validated using data collected from the Yeongjong Bridge in Incheon, Korea.