Tag Archives: infrastructure

I-35W ‘Smart Bridge’ Test Site Uses Vibration Data to Detect Bridge Defects

By analyzing vibration data from the I-35W St. Anthony Falls Bridge, MnDOT is working to develop monitoring systems that could detect structural defects early on and ultimately allow engineers to improve bridge designs.

“With data spanning several years, the I-35W St. Anthony Falls Bridge offers a unique opportunity for investigating the environmental effects on a new concrete bridge in a location with weather extremes,” said Lauren Linderman, Assistant Professor, University of Minnesota Department of Civil, Environmental and Geo-Engineering. Linderman served as the research project’s principal investigator.

“This project gets MnDOT closer to using bridge monitoring systems in combination with visual inspection to help detect structural problems before they affect safety or require expensive repairs,” said Benjamin Jilk, Principal Engineer, MnDOT Bridge Office. Jilk served as the research project’s technical liaison.

2017-01-bridge.png
Completed in 2008, the I-35W St. Anthony Falls Bridge has a smart bridge monitoring system that includes hundreds of sensors.

What Was the Need?

In September 2008, the I-35W St. Anthony Falls Bridge was constructed to include a “smart bridge” electronic monitoring system. This system includes more than 500 sensors that continuously provide data on how the concrete structure bends and deforms in response to traffic loads, wind and temperature changes. Transportation agencies are increasingly interested in such systems. As a complement to regular inspections, they can help detect problems early on, before the problems require expensive repairs or lead to catastrophic failure. Smart bridge systems can also help engineers improve future bridge designs.

The smart bridge system on the I-35W St. Anthony Falls Bridge includes accelerometers, which provide data on the way the bridge vibrates in response to various stimuli, including structural damage. Vibration-based monitoring has the advantage of allowing damage to be detected at any location within the bridge rather than only at the specific locations where measuring devices have been placed.

However, it can be difficult to use vibration monitoring to detect damage when vibration is masked by the bridge’s natural response to traffic loads, wind, temperature changes and other environmental conditions. A crack in a bridge girder, for example, can produce a vibration signature similar to one produced by a change in beam length due to variations in temperature or other causes. Consequently, since 2008 MnDOT has conducted a series of projects using data from the St. Anthony Falls Bridge to establish a way to distinguish anomalous data indicating a structural defect or damage from background “noise” associated with other causes.

What Was Our Goal?

This project sought to develop a method for analyzing accelerometer data from the I-35W St. Anthony Falls Bridge that would show how the bridge naturally vibrates due to traffic, wind and other environmental conditions. With this fingerprint of the bridge’s natural vibration, engineers would have a baseline against which to measure anomalies in the data that might indicate structural damage.

What Did We Do?

A large amount of data has been collected from the bridge since its construction. To establish the vibratory fingerprint for the bridge, researchers examined the frequencies and shapes (or modes) of bridge vibration waves. The method they used to identify the data segments needed for the fingerprint was to evaluate the peak amplitude of bridge vibration waves and their root mean square (RMS), a measure of the intensity of free vibration.

The researchers applied this method to the vibration data collected on the I-35W St. Anthony Falls Bridge between April 2010 and July 2015, calculating the average frequencies for four wave modes and determining how they varied with the bridge’s temperature. They also calculated the way frequencies changed with the bridge’s thermal gradients, or variations in temperature between parts of the structure.

What Did We Learn?

The methods developed in this project were successful in establishing a fingerprint for the way the I-35W St. Anthony Falls Bridge vibrates due to environmental conditions, and a way to evaluate changes in vibration over time indicative of structural damage or other factors.

Researchers found that the ratio of peak signal amplitude to RMS in bridge vibrations was a strong indicator of data that should be analyzed, and was evidence of a large excitation followed by free vibration. By themselves, peak amplitude and RMS cannot distinguish between ambient free vibration and forced vibration.

Researchers were able to use this method to successfully analyze 29,333 data segments from the I-35W St. Anthony Falls Bridge. This analysis revealed that as temperature increases, the natural frequency of vibration tends to decrease. The magnitude of this change, they concluded, must be related not just to the elasticity of the bridge but also to other factors such as humidity. However, temperature gradients within the bridge did not appear to have a significant effect on the natural frequencies of the structure.

What’s Next?

MnDOT will continue to collect data from the bridge as it ages to further understand its behavior. This will provide an opportunity to determine how anomalies in vibration data correspond to cracking and other forms of structural distress. Ultimately, MnDOT hopes to use this bridge monitoring system in combination with visual inspection both to detect problems in bridges earlier and to develop better bridge designs. Researchers are also currently working on a follow-up project, Displacement Monitoring of I-35W Bridge with Current Vibration-Based System, to determine the effects of temperature on the bridge’s dynamic and long-term vertical displacements, which can be used to monitor the bridge’s stiffness, connections and foundations.

—-

This post pertains to Report 2017-01, Feasibility of Vibration-Based Long-Term Bridge Monitoring Using the I-35W St. Anthony Falls Bridge, published January 2017. 

Transportation spending: How does Minnesota compare with other states?

Transportation funding continues to be a contentious issue in Minnesota: Are we spending enough, too little, too much? One way to help answer that question is to compare spending with other states.

“A simple comparison, however, may not accurately reflect the real level of transportation funding across the states,” says Jerry Zhao, an associate professor in the Humphrey School of Public Affairs. “States face different levels of demand and costs due to different geographic, demographic, or labor market conditions.”

To better understand the factors that influence the transportation funding level, Zhao and Professor Wen Wang at Rutgers University developed a cost-adjusted approach to systematically compare highway expenses among states. They found that while Minnesota spends more than average on highways, its spending level actually ranks low in cost-adjusted measures.

“We controlled for the effects of some major cost factors, such as demographics and natural weather conditions, which are outside of the control of state and local officials,” Zhao explains. “We found that natural weather conditions have a significant impact on highway spending—a lower winter temperature is associated with higher highway expenditures.”

The effect of population size isn’t as straightforward: “There is some impact of economy of scale, but only to a certain threshold,” he says. While urban areas have greater complexity, the higher population density is associated with less spending per capita, probably due to spreading the costs across a greater population.

The analysis also found that state and local governments tend to spend less on highways when they are under fiscal stress, and states with a higher gross domestic product (GDP) appeared to spend more on highways per capita. “Essentially, highway investment decisions may be greatly influenced by the economic fluctuations and fiscal stresses faced by a state,” he says.

According to unadjusted 2010 data, Minnesota ranks 8th on highway spending per capita and 18th on its share of statewide highway spending in GDP. “But after adjusting for those factors that are largely out of control by transportation policy, we found that Minnesota’s rankings drop to 37th on highway spending per capita and 41st on the share of highway spending in GDP,” Zhao says. “This suggests that the relatively high level of highway spending in Minnesota is largely driven by the cost factors of demographics and weather conditions.”

“This study confirms what MnDOT has experienced and that transportation financing is more complicated than one would expect,” says Tracy Hatch, MnDOT deputy commissioner. “Not only is Minnesota’s transportation system significantly undercapitalized—there are considerable financial impacts from factors outside of our control.”

The analysis was conducted as part of the U’s Transportation Policy and Economic Competitiveness Program (TPEC). In previous work, TPEC researchers created the Minnesota Transportation Finance Database, which compiles data about Minnesota’s transportation finance and shows the change of transportation spending in Minnesota over time.

Thicker may not equal stronger when building concrete roadways

Transportation agencies have long placed high importance on the thickness of their concrete roadways, making it a major focus of control and inspection during construction. While it is commonly believed thicker concrete pavements last longer, there is little data to support this claim.

“One big reason for the lack of data on the relationship between concrete pavement thickness and performance is the destructive nature of these measurements,” says Lev Khazanovich, a former professor in the University of Minnesota’s Department of Civil, Environmental, and Geo- Engineering. “Concrete thickness is typically assessed by coring—a destructive, expensive, and time-consuming test that only offers widely spaced measurements of thickness.”

In a MnDOT-funded study, U of M researchers set out to fill this knowledge void by leveraging recent advances in the nondestructive testing of pavements that allow for large-scale, rapid collection of reliable measurements for pavement thickness and strength. They conducted four evaluations on three roadways in Minnesota using ultrasonic technology to collect more than 8,000 measurements in a dense survey pattern along with a continuous survey of observable distress.

“We found that both pavement thickness and stress measurements are highly variable, with a half-inch of variation in thickness about every 10 feet,” Khazanovich says. “Interestingly, three of the four surveys averaged less than design thickness, which is contrary to typical accounts of contractors building slightly thicker slabs in order to avoid compensation deductions.”

Data analysis showed that exceeding design thickness did not seem to increase or decrease pavement performance. However, a measurement of pavement strength and quality known as “shear wave velocity” did produce valuable findings. “A drop in the shear wave velocity strength measurement corresponded to an increase in observable pavement distresses such as cracking and crumbling,” Khazanovich explains. “This was especially apparent when we were able to easily identify locations of construction changes, where significant changes in shear wave velocity matched up with observable distress.”

The results of this study illustrate the importance of material quality control and uniformity during construction, since alterations in pavement strength and quality may significantly influence pavement performance. In addition, researchers say that despite inconclusive thickness results, it is still important that pavement has significant thickness to carry its intended traffic load over its service life. Finally, the study demonstrates that new methods of ultrasonic shear wave velocity testing are useful for identifying changes in construction and design that could lead to higher rates of pavement distress.