Troubleshooting from Afar

Perspectives: Research and Creative Activities at SIUC, Spring 2008

:: research survey ::

Troubleshooting from Afar

The city water meter reader's job got easier a couple years ago as wireless sensors made it possible to gather the readings from a remote location, such as a vehicle driving down the street.

But that remote data collection technology can't hold a candle to what researchers at SIUC are working to create.

Engineers and scientists from several fields, working through SIUC's Materials Technology Center, recently received $1 million from the Federal Highway Administration's Intelligent Transportation Systems program to create a network of Internet-based wireless sensors and databases that can provide near real-time information about the structural soundness of transportation infrastructure, such as a bridge.

Shing-Chung "Max" Yen, the center's director, says the system will combine an integrated sensor network with wireless technology and powerful data analysis tools that potentially would improve transportation safety and efficiency. Engineers also could apply similar systems to other situations, such as security systems and environmental monitoring.

"You want to get this information quick, even when you are away from the area," Yen says. "If we develop this technology the right way, it will benefit many areas. These are global concepts and there are many things we can do."

The three-year project, which the University is also committing $1 million to in terms of faculty release time and other subsidies, will involve several faculty from various engineering and science-related departments, including civil and environmental engineering, chemistry and biochemistry, and electrical engineering. Undergraduate and graduate students also are playing a role in the research.

The team first will focus on creating and applying the system to a footbridge on campus to test ideas. Eventually they will work with the Illinois Department of Transportation to wire an actual vehicle bridge somewhere in Southern Illinois.

The idea involves placing different kinds of sensors that can detect variables such as stresses, elongation, deformation, and rotation, at key points on the structure. The researchers will connect the sensors to wireless transmitters, each set up as their own Internet address, which will then send data to a server off-site. The researchers will download that data to a Web site for analysis.

The sensors must not only transmit data about how the structure is behaving, but must also crosscheck each other to ensure the data is reliable. For example, the sensors should show the effects of damage to a bridge at the point of damage, but sensors farther away might also validate those effects by signaling that the bridge's structure is compensating for that damage elsewhere.

Only then do the sensors become a true network and really begin telling the story of the structure's health, creating what Yen calls the structure's "signature."

Some of the sensors may use nanoelectronics to pick up minute signs of deterioration, allowing engineers to prevent a much bigger problem down the line.

One of the biggest challenges the researchers face, however, is fast, accurate analysis of what promises to be a huge amount of incoming data. Without that, the data is mostly useless.

A sizeable portion of the researchers' work, therefore, will go into compiling databases incorporating various parameters against which they can compare the new, incoming data from sensors. Because such databases are largely nonexistent, the researchers will build their own. They will use data they can apply from existing sets—such as corrosion data from the auto industry—along with their own predictions about how a particular bridge structure should react under different circumstances, based on past experience, actual examples, and computer models.

Such databases might include scenarios for how a bridge might react if it is damaged, if it receives faulty repairs that change its dynamics, or how those dynamics change as the bridge simply gets older.

"You have to be knowledgeable enough to know what your data means," Yen says. "You have to be able to tell what is reliable data and what is not. So you have to have a lot of databases.

"If an emergency comes up, we're looking for quick assessments. We don't have time to do computations. If you can do [quick assessments], there is still a chance to minimize the potential catastrophe."

Yen foresees a time when many bridges have their own performance record database on file that can be used, updated, and consulted as the structure ages and undergoes repairs and modifications. Depending on how successful the team is, Yen says the sensor network coupled with the powerful, layered databases will allow for much quicker diagnosis of structural problems with a bridge than civil engineers using traditional methods can perform. At the very least, the system would better indicate when a thorough, traditional evaluation is called for, making bridge inspections more efficient overall.

Eventually, such analysis may result in more efficient bridge design. Bridges are sometimes overbuilt to ensure safety over the estimated life of the structure. However, an overbuilt bridge may not be safer in the long run, says Yen. The SIUC research may indicate better ways to build safer, longer-lasting bridges.

Infrastructure is simply the selected proving ground for the new networked sensor system. It also could be applied to many other uses.

Security is an obvious possibility. Such a network could work in concert with ultra-sensitive sensors that "sniff" the air for chemicals emitted by methamphetamine labs or explosives, keeping the public and law enforcement out of harm's way. Ling Zang, associate professor of chemistry and biochemistry at SIUC, is working on such sensors and he will work on the transportation project as well, Yen says.

Such a system also could be installed in environmentally sensitive areas, such as in or around public water supplies, to give constant remote feedback and analysis and quickly pinpoint and warn of any contamination.

SIUC researchers have already developed several components of the planned system: a displacement sensor and a corrosion sensor, a data acquisition system to translate materials information (about deformation, for example) into digital format for transmission, and several radio-frequency–based communication systems.

Yen hopes the project also will help SIUC maintain an updated curriculum for engineering students, keeping them well prepared for the workplace after graduation.

"This project will allow us to exercise a real job situation and can help us influence the potential placement of our students," he says. "It will also help us with our curriculum. The engineering industry is changing and we also need to be changing."

—by Tim Crosby