By: Dr. Toader A. Balan, Peter Vanderzee and Frank Wingate, P.E.
Owners and engineers have started using structural monitoring technology as a means to provide precise, objective and timely information to improve the management of bridges and other civil structures.
However, there are other powerful, commercially available analytical tools, when coupled with structural monitoring, that will fundamentally change the practice of structural engineering and long-term bridge management.
The financial impact from use of these advanced analytical tools has the potential to save billions of taxpayer dollars, greatly enhance public safety and confirm a better overall condition rating for this nation’s infrastructure than visual inspection protocols have shown.
Nice to know is now need to know
Structural monitoring as a stand-alone technology was conceived in academia nearly 20 years ago. By creatively combining unique technology platforms (e.g., highly accurate sensing devices and improved computing power) proponents demonstrated how structure owners could benefit from previously unavailable objective knowledge about structural condition and performance under load. Early systems typically captured voluminous data over short time periods (days or weeks) in locally mounted data-collection devices (data loggers) for subsequent analysis by professors and their graduate students. Written reports were provided to owners, but much of this early effort was characterized by owners as “nice-to-know” information instead of being used to actively manage deficient structures or structures with known defects over an extended time period. What owners needed and wanted was objective, timely information that could be used to support difficult decisions regarding maintenance, repair or replacement.
Commercial firms started offering improved structural monitoring technology and better decision-making information approximately 10 years ago. What differentiated commercial offerings from the earlier academic approaches was the integration of wireless data transmission and presentation of captured data over the Internet. These enhancements made it possible to conduct near-real-time management of deficient structures, since the information was available on demand instead of the standard biennial, subjective visual inspection report mandated by the Federal Highway Administration (FHWA).
While bridge owners have generally been cautious to adopt commercial structural monitoring technology, this technology should now be considered “standard of care” for professional engineers, especially for use on structures that are classified as deficient. The technology has been widely demonstrated, is commercially available from a number of firms around the world, and is even being specified for use on new structures (e.g., the I-35W replacement and the Q Bridge in Connecticut). In a legal context, it may soon be more difficult for owners to explain why they decided not to use structural monitoring technology than to aggressively use the technology on bridges that need more objective, timely information for long-term bridge management.
A multi-million-dollar decision for bridge repair or replacement that is not sufficiently supported by accurate, reliable information makes the long-term bridge management process financially less efficient. For example, a visual assessment of section loss (corrosion) may not be sufficiently reliable to formulate an effective repair program. Knowing the actual distribution of strain/stress in a structure with visible section loss, an engineer can more confidently quantify the effects of live-load, temperature and bearing performance, which may be quite different than the initial design assumptions or visual observations. Using enhanced analytical tools, the engineer can pinpoint locations of the highest strain/stress, allowing more efficient inspection protocols, improved operational safety and safe extension of life span, all while driving lower life-cycle costs. Owners, users and the taxpayers all win with better information.
The over-under
Projects with early adopters provided a significant amount of practical application experience for solution providers, structure owners and engineering consultants. Owners who were early adopters of structural monitoring technology gained valuable first-hand experience and now stand ready to implement even more sophisticated applications, such as the integration of monitoring with finite-element modeling (FEM). However, integration of monitoring and modeling, while capable of providing a robust technical platform for enhanced long-term bridge management, should only be attempted by technical teams with strong instrumentation and structural analysis credentials. The inclusion of a highly experienced structural engineering consultant is crucial for success.
Significant field experience has been invaluable for making practical recommendations to structure owners on how to get the highest return from their investments in structural monitoring solutions. For example, we offer the following observations, reached after years of relevant field and analytical experience:
- Data intensity is overrated. Instead of collecting data at 1,000 points per second (Hertz), adequate data streams to support detailed analysis of structural condition can be as low as 6-10 data points per day per sensor over an extended time period;
- Sensor accuracy is overrated. Owners and engineering practitioners do not make different decisions if a structural member exhibits one microstrain more or less than expected. Adequate sensor accuracy on the order of ±30 microstrain will support difficult decisions;
- Monitoring periods are underrated. Unless structural data is collected through a full ambient temperature cycle (nine to 24 months), critically important data needed to diagnose structural deficiencies will be missed. In addition, this observation puts a premium on the use of rugged sensors and professional installation (e.g., conduit, reliable power supplies, antennas) that can provide years of uninterrupted service;
- The number of sensors is less important than their proper operation.
- By using sensors that can capture both tensile and compressive displacements, the owner can minimize the number of sensors and overall solution cost. The oft-posed question about the need for exact placement of sensors over high-stress areas is a canard. We have proven that cost-effective structural analysis, including sophisticated FEM, does not require hundreds of sensors to achieve good results; and
- A professionally managed network operations center (NOC) is crucial. The “server-in-a-closet” approach to capturing data from critical structures is not worthy of owner consideration when managing critical structures over long time periods, especially when structures are already classified as deficient. Imagine the liability generated when a structure fails because the owner did not get crucial data in a timely manner after the server-in-a-closet crashed while the solution provider was on vacation. Owners should query solution providers about data security, NOC uptime statistics and the ability to provide timely alerts when predetermined displacement or strain limits have been exceeded.
While the previous observations represent practical advice for owners and engineers seeking to employ advanced structural monitoring technology, the main purpose of this article is to inform readers about the benefits of using the next generation of condition-assessment technology that can dramatically improve long-term bridge management. After considering the previous observations, readers should be asking the following question: “If I have long-term data from two dozen sensors on my long-span bridge over a full ambient temperature cycle, how do I use this data to maximize the life span of my bridge, expand the safe operating envelope, reduce liability and reduce life-cycle costs?” The answer is to use the captured monitoring data to build a calibrated FEM of the structure.
Of your concern
Each FEM is different. That is a critically important statement supported by first-hand experience and software modeling capability. While it may be easy for the reader to recognize the benefits of a 3-D model over earlier 2-D models, there are other crucial concerns that must be considered when using an FEM for structural analysis. Here are a few:
- If the FEM-predicted structural-response parameters (displacements, strains, frequencies, etc.) and the recorded data do not agree within reasonable limits, the model is virtually useless. At the same time, a calibrated FEM can be used to verify if instrumentation devices are properly installed and functioning as expected;
- Model assumptions and restrictions (e.g., boundary conditions, material and section properties, adopted finite element mesh and type of finite element used [e.g., thick shell or thin shell]) must all be open to modification in order to properly calibrate the FEM;
- The FEM should be capable of near-real-time automatic updating, especially if the superstructure is in very poor condition (e.g., fractures due to crack initiation and propagation, out-of-plane bending). This capability is not a trivial exercise; and
- The calibrated FEM should be utilized frequently to improve the long-term management of the structure:
-
- To conduct “what-if” analysis for temporary loading scenarios (thermal gradient loads, wind loads, permit live loads, temporary loads, etc.);
- To identify the high stress concentration areas or local damage zones for heightened (guided) inspection and enhanced monitoring;
- To evaluate and confirm actual structural condition instead of relying solely on subjective, biennial visual inspections;
- To identify other locations or parameters that should be studied to further optimize long-term bridge management; and
- To optimize relocation of sensors or the addition of sensors to capture the most valuable information possible.
The development and use of FEMs for long-term bridge management is a job for highly experienced structural engineers. To properly build a calibrated FEM and conduct professional structural analysis takes significant engineering experience and a wealth of conceptual and practical information about structural behavior.
Differences in modeling software matter a great deal. While a number of different FEM programs are commercially available, some are more comprehensive and capable of better formulation (accuracy) than others. Engineering consultants and owners should be careful in the selection of modeling software and doubly careful in evaluating the experience of a modeling practitioner before using this technology to manage bridge structures worth hundreds of millions of dollars.
Modern structural-modeling software allows consideration of the following details: structure specifics (precise connectivity of beam and shell elements by use of rigid links, beam-end offsets, beam/shell-end releases, linear/nonlinear fixities, etc.), different types of nodal/element loads, different loading scenarios during construction and for post-construction verification as well. In this article there are illustrated examples of FEM capabilities used for integrating structural modeling and monitoring technologies.
The first example on p 44, generated by use of MIDAS/Civil [1], depicts an FEM used for monitoring and modeling of an arched bridge.
It is important to note that strain/stress in each of the individual elements can be determined by using a calibrated FEM.
By knowing versus assuming the distribution of strain/stress, engineers can develop and implement much-improved long-term bridge-management plans. And improved bridge-management planning is the first step to safely extending the life span of a bridge and lowering life-cycle costs.
In an era of insufficient federal and local transportation funding, knowing is clearly better than assuming to financially optimize long-term bridge management.
The second example on p 46, created by use of MIDAS FEA [2], shows an FEM used for monitoring a deck truss bridge.
Integration of structural monitoring technology and modern FEM can provide the following benefits for structural engineering practitioners as they seek better methods to optimize long-term bridge management:
- Evaluation of structural performance under applied loads (including thermal) for use in determining actual condition and long-term management of known defects;
- Facilitating the planning of inspection and maintenance actions to maximize effectiveness and minimize costs;
- Validating design assumptions and structural parameters (span spacing, material and section properties, etc.), especially for major structures that undergo rehabilitation as part of a preservation program;
- Eventually, the ability to predict the remaining life span of a structure, especially important with tight budgets, balanced against the need for funding expanded transportation capacity; and
- Updating and revising design manuals and standard specifications.
More brains than money
When properly developed and utilized, calibrated FEMs can reduce liability exposure, maximize structural life span, lower life-cycle costs and expand the safety envelope for older, deficient structures. The use of monitoring and modeling technology requires economic justification to be sure, but for any contemplated major action (repair, replacement) over $2 million, it should be considered by structure owners.
However, knowledge and experience in the use of these technologies is crucial. Inadequate monitoring and modeling that lead to inappropriate decisions are not worth any price, even if they are free.
In summary, we believe that optimizing long-term management of large, critically important civil structures will depend upon use of both monitoring and modern analytical tools—the synergistic integration of structural monitoring and FEM.
The authors would like to acknowledge the contributions of MIDAS Information Technology Co. Ltd. of Seoul, South Korea.
About The Author: Balan is a professor of structural engineering and a technical consultant with Michael Baker Jr. Inc. Vanderzee is president and CEO of LifeSpan Technologies Inc. and Wingate is a chief engineer at LifeSpan Technologies Inc.