Skip Header NavigationIntranet 
CENTER FOR EMBEDDED NETWORKED SENSINGContactDirectionsEmploymentEventsNews
HomeAbout UsResearchEducationResourcesPeople

Research Project


Structural Health Monitoring and Assessment: Tall and Special Buildings

Applications > Seismic > Structural Health Monitoring and Assessment: Tall and Special Buildings

On this page: Overview | Approach | Accomplishments | Future Directions | People

Lead Investigators

John Wallace, Bill Kaiser

Overview

This research envisions using the greater Los Angeles infrastructure of tall and special buildings (Fig. 1a) to develop and implement a fully-functional embedded sensing network for structural health monitoring and damage assessment (Fig. 1b, 1c, 1d). The resulting super network of monitored buildings envisioned (Fig. 1d), embedded with innovative, model-based sensor deployments, wireless sensing technology, and damage assessment and visualization tools (Fig. 1c, 1d), will provide a novel environment to address critical issues related to structural analysis and performance-based design of buildings. Key components to the success of the proposed research are shown in Fig. 1(b), and include collaboration with industry and educational partners and the development of a user-friendly html-based data visualization portal. The long-term vision of the project, if additional resources are garnered, are to: (2) enable rapid emergency response and post-event assessment and (3) improve loss estimation scenarios using tools such as HAZUS, (4) develop unique education and interaction opportunities for K-12, undergraduate, and graduate students, and (5) establish outreach activities that target the general public, public policy makers, real estate developers, and research/practicing engineers. Three NSF proposals were submitted, one to NEESR-SG, one to CMMI Sensors & Sensor Innovation, and one to NSF PIRE in an attempt to secure additional funding to support these long-term goals. 

Figure 1

This research will provide critically needed field data to address fundamental issues associated with modeling of complex structural systems, including system interactions. These data cannot be generated in controlled laboratory experiments due to issues of scale, materials, boundary conditions, system interactions, and most significantly, cost. The research vision is to make the true built environment the laboratory – to collect the wealth and diversity of data not otherwise possible. The project will provide vital data to improve modeling capabilities for wind events, low-to-mid level amplitude shaking in relatively frequent earthquakes, and ultimately, the “Big One”. The systems and tools will be developed in an open-source environment. The project team, along with the NEES@UCLA and UCLA CENS resources and technologies, are uniquely qualified to carry out the proposed research.

The primary goal of this research is the development of a robust SHM system, along with the associated hardware and software, using tall and special buildings in Los Angeles as a testbed. To achieve this goal, the proposed project is divided into three task groups; (1) a structures group to implement and provide the application testbed, (2) a sensors group to develop/integrate a suite of new/existing sensors, and (3) the network group charged with developing a toolbox of required hardware/software; SHMBox. These three groups will interact with industrial partners and a scientific advisory board to achieve the vision of the proposed research. The final component is the proposed educational program, which is discussed later. The organization of the proposed research and the primary research tasks of each group is illustrated in Fig 1b.

Approach

Prototype SHMBox: Our initial task involves configuring a prototype SHMBox using existing sensor transducers. A small network of 2 to 3 prototype SHMBoxes will be tested in the laboratory using a small scale structure and nees@UCLA linear shaker, to allow for side-by-side comparisons of both wired and wireless systems for a broad array of structural configurations, response quantities, and response amplitudes. Iteration will be used to develop a field deployable box.

Additionally, we will test and continue development of the novel RBS time synchronization scheme initially developed by CENS (Elson 2001, Ganeriwal 2003). The addition of this feature will overcome problems that have been observed with GPS time-stamping (e.g., nees@UCLA Four Seasons forced vibration testing project, see Yu et al, 8NCEE, 2006; and the CENS-Seismic Group Mexico deployment). Another group task will be the integration work required to ensure that various sensor types (e.g., pressure, load, acceleration, velocity, displacement, and strain) can be interfaced with the SHMBox., This is an important task to achieve the vision of a robust system for rapid deployment (i.e., “plug-and-play” for users) Although a wired network avoids the problems associated with GPS time-synchronization, it is cumbersome and expensive due to the extensive installation process (e.g., based on experience from the Four Seasons building test and the UCLA Factor building system upgrade).

Ongoing work for the software deployment tools in the GeoNet project will be extended to include the SHM systems. In particular the exploration of 802.11 and 802.15.4 radio characteristics and capabilities in urban environments for the development of software deployment tools will be extended to explore the envisioned building scenarios 

Deployment and Iterative Improvements: Upon satisfactory completion of prototype SHMBox development, initial field deployment will take place in an instrumented building with wired accelerometers (e.g., UCLA Factor Building). The Factor building provides an excellent location for an initial test deployment (http://www.seas.ucla.edu/~skolnik/). It has an extensive wired embedded network consisting of 72 uniaxial accelerometers, nine state-of-art data loggers, fiber optical network cables, and several nearby borehole seismometers buried up to 300ft deep. The network is managed by CENS seismic team (Skolnik 2006). Results of the initial field study will be used to assess and/or redesign the prototype system to improve important features (functionality, toughness, etc). It is essential that the proposed system be easy to install and include remote assessment of the sensor and sensor network health. A major drawback with current implementations (including CSMIP and LA-DBS) is that they often are inoperable, generally unknown to the public, and lack consistency (e.g., data loss). Once a robust system is developed, field deployments of multiple networked sensor groups will allow for various SHM deployments (sensor types, locations, communication, etc) to further develop the system capabilities. We envision event detection to be associated with a threshold response feature (sensor or group of sensors) specific to each building, and this task will be studied both in the laboratory and in the field.

Once the initial field deployment work is completed, and if we secure additional funding, we will work with our IAB and our SAB to identify 3 or 4 promising buildings for field study, either existing buildings or buildings under construction. In this phase of the study, we expect that our systems would supplement the required LA-DBS systems, and they would probably be temporary. With the proposed system, we expect to install an 18 to 24 node network in a day, with remote, real-time data display available at the end of the same day, initially using SensorBase.org (this has already been accomplished for environmental sensor deployments). Further, a user-friendly, menu driven, web-based template will be developed to enable users to define specific attributes for a given SHM deployment (interactive drawings with sensor locations, types, calibration factors, etc). The eventual goal of the proposed research will be to create tools to ingest data for events of interest to the NEES data repository.

Data Manipulation & Visualization: Within the overall context of the research plan, a current stumbling block for SHM implementations is the numerous cumbersome programs involved in the overall process of data collection and assessment (for each sensor as well as the overall system). An early attempt to consolidate data from a network of buildings is the CSMIP-3DV program (Naeim 2006). Rapid and easy assessment of damage is the ultimate selling point of SHM systems to building owners, researchers, and decision makers. In the proposed research, we will collaborate with Naeim to complement the existing tool (CSMIP-3DV) by implementing novel tools for near real-time data visualization and for automated system identification. We have already explored this option, established collaboration, and initiated preliminary work to explore the most productive approaches to accomplish these goals (i.e., using the NEES@UCLA linear shaker and a small-scale model structure shown in Fig. 7).

Super Network:  The issue of managing the emerging super-network of monitored buildings (e.g., where to send/store data and who is responsible for data management and maintenance) will be assessed to develop recommendations for future work in the event that additional funding is obtained. For example, in the NEESR Solicitation requirements, data sets generated by the proposed research will be uploaded to the NEESit data repository; however, longer term solutions for the wealth of data that would be generated by the building network will be assessed. Options include continued use of the NEESit infrastructure, or possible the CENS html-based data center SensorBase.org infrastructure, or some other option. Regardless, the ultimate goal of the proposed research would be to enable a command center (initially located at UCLA, but ultimately virtual) that could be used to display the results of the damage and performance assessment for each building of the limited, initial network (3 or 4 buildings). In the long-term, a promising approach that will be explored to create a super network (tens or hundreds of buildings) would be for LA-DBS to modify their permit process to require participation in the network. The results of the SHM for the network could be displayed on a map and for each building concise damage state probabilities (e.g., Figure 1d). The rapid dissemination of this type of information would enable informed allocation of resources associated with emergency response in extreme events. Lastly, aggregate damage measures developed from the individual building statistics for a given region could be generated (or even predicted for hypothetical events to inform emergency planning and response. Note that current HAZUS scenario events are based on idealized modeling (SDOF) and associated fragility relations (i.e., pre-1973 concrete frame), and one sub-task of the proposed research plan will investigate potential improvements in HAZUS.

Potential Collaboration: Collaboration has already been initiated with the LA Department of Building and Safety (Nick Delli Quadri), the California Strong Motion Instrumentation Program (Tony Shakal), and USGS (Woody Savage). The primary goals of this collaboration are to: (1) conduct pilot studies in actual buildings, (2) revise the LA-DBS instrumentation requirements, and (3) influence USGS/ANSS instrumentation programs. 

NEES@UCLA Enhancements: The study will result in a number of significant enhancements to the NEES@UCLA Equipment Site. These enhancements include: (1) new sensor types and interfaces, (2) new (and presumably cheaper) dataloggers and data acquisition software alternatives to the Kinemetrics Antelope system (we do not envision replacing Antelope at this time, but providing an alternative approach), (3) vastly improved monitoring abilities (local storage, wireless data transmission), and (4) improved data visualization for field testing. These enhancements are very significant, and are only possible due to technology advances that have been made in recent years.

Accomplishments

Future Directions

People

Faculty

Researchers:

Staff:

Graduate Students:

External Research Partnerships