Solar-Powered IoT Sensors Could Improve Nation's Infrastructure
Tom Lombardo posted on October 13, 2019 |

An analysis of the U.S. Department of Transportation’s 2018 National Bridge Inventory (NBI) database suggests that roughly 9 percent of all U.S. bridges are considered “structurally deficient.” That's more than 54,000 potentially unsafe bridges. Even worse, the rate of repair is lower now than it was five years ago. Think about that the next time you cross a bridge. (I do.)

Figure 1. A bridge in Florida outfitted with sensors. (Image courtesy of the National Science Foundation.)
Figure 1. A bridge in Florida outfitted with sensors. (Image courtesy of the National Science Foundation.)

The good news is that the growth of low-power smart sensors, the Internet of Things, solar power, and battery technology could help engineers detect significant problems before they become catastrophic. One researcher working to make that happen is Dr. Jennifer Bridge, professor of civil engineering at the University of Florida.

Assisted by a grant from the National Science Foundation, Bridge and her students are developing a sensor network that monitors the status of bridges during tropical storms and reports that information in real time. The initial deployment focuses on bridges that are integral in evacuation and rescue operations, especially those that are similar to bridges that suffered structural damage from previous hurricanes. As the price of this technology continues to fall, we could see a day when all bridges contain smart sensors that constantly monitor the stresses on bridges and alert civil engineers of impending failures.

Figure 2. Sensor hub. (Image courtesy of the National Science Foundation.)
Figure 2. Sensor hub. (Image courtesy of the National Science Foundation.)

Here’s a short video describing the sensor network:

(Video courtesy of the National Science Foundation.)

After watching the video, I contacted Bridge, who was kind enough to answer some questions. Our exchange follows:

TL: I got a glimpse of the hub’s internals in the video. Is that an Arduino board controlling the system? Raspberry Pi?

JB: Texas Instruments Microcontoller LaunchPad (similar to the systems you mention).

TL: How much data can the hub store in the event of a communication loss?

JB: We are currently using a 32 GB SD card. This will hold a few weeks of data at our current sampling rate and data collection duty cycle.

TL: The data link between the hub and the university—is that a cellular link?

JB: Yes. We are using a Digi XBee Cellular Modem, LTE CAT 1 (currently using the AT&T version but plan to switch soon to the Verizon version at this site due to better coverage in the area). These are very small form factor, relatively low power cellular modem options that allow us to run on solar/battery power while streaming data in real time from the seven pressure sensors. The trade-off is lower data throughput than larger cellular modems (such as Cradlepoint or Cisco routers). Other sites we monitor have relied on larger, more powerful modems, but we have had access to Mains power that made this possible. This particular site has no power in the vicinity (and we don’t want to rely on power during storm events), so we rely on solar/battery.

TL: How do you make the hub capable of withstanding high winds and rain?

JB: The hub has a secure mounting to the railing of the bridge to ensure it stays in place in strong winds. The enclosure has an IP rating of 67 or higher. We do make modifications to the enclosure to allow the connection of the sensor and solar power components. These holes are sealed using a marine rated sealant, such as 3M Marine Seal. 

TL: I see from the video that you determine lateral forces by measuring the pressure difference between the two sides of the bridge. You mentioned that the next step is to measure the response of the structure itself. What type of sensors would you be using for that, and what potential responses do you anticipate?

JB: Accelerometers can provide a measure of the vibration response of the structure, which can change as it undergoes damage. We can also use sensors that more directly measure response, such as tiltmeters (to measure tilt) and extensometers/displacement transducers (to measure bridge deck unseating/uplift).

TL: In what ways can your research and the underlying technology be used to help maintain the general infrastructure of the U.S.?

JB: Improving infrastructure relies both on maintaining our existing inventory of structures and designing more resilient new structures. Structural response data can, in some cases, provide an early warning of the onset of damage that can supplement and complement existing visual inspection techniques. Understanding how structures respond to real loading conditions enables us to make more informed decisions on when maintenance and repair is needed and most effective and what types of retrofits are most appropriate. In addition, when we better understand the loading conditions experienced by structures, we can improve how they are designed to respond, thereby improving their resiliency and increasing their life span. Wave loading design for bridges currently relies on equations derived from laboratory experimental testing and analytical models; full-scale measurements during storm events are required to better understand this type of loading.

As the powers that be continue to neglect the nation’s infrastructure, it’s good to know that some engineers are working to ensure its safety and reliability.


_______________________________________________________________


Follow Tom Lombardo on Twitter, LinkedIn and Facebook.



Recommended For You