A smartphone app can check the structural integrity of bridges, research shows
There may be a smartphone application for knowing whether San Francisco’s Golden Gate Bridge, or any other bridge for that matter, is in good shape, according to a study.
New research shows that mobile phones installed in vehicles, equipped with special software, can collect useful data on structural integrity while crossing bridges. In doing so, they can be a cost-effective alternative to sensor sets attached to the bridges themselves.
“An important finding is that information about the structural health of bridges can be extracted from accelerometer data collected by smartphone,” said Carlo Ratti, co-author of the study.
The study was conducted, in part, on the Golden Gate Bridge itself. The study, involving researchers from the Massachusetts Institute of Technology (MIT), US, showed that mobile devices can capture the same type of information about bridge vibrations combined with stationary sensors.
Researchers also estimate that depending on the age of the road bridge, mobile device monitoring can add from 15 to 30 percent more years to the life of the structure.
“These results suggest that large and inexpensive data sets collected by smartphones can play an important role in monitoring the health of existing transportation infrastructure,” the authors wrote in their new paper, published in Nature Communications Engineering.
Bridges vibrate naturally, and to read the important “model frequency” of that vibration in many places, engineers often place sensors, such as accelerometers, on the bridges themselves.
Changes in modal frequencies over time may indicate changes in the integrity of the bridge structure.
To conduct the research, the researchers developed an Android-based mobile application to collect accelerometer data when the devices are installed in vehicles passing over the bridge.
They then saw how well that data matched data recorded by sensors on the bridges themselves, to see if the cell phone method worked.
“In our work, we designed a method to extract modal vibration frequencies from noise data collected from smartphones,” said lead researcher Paolo Santi.
“As data from multiple trips over the bridge are recorded, the noise created by the engine, suspension and vibration of the traffic, (and) the asphalt, are often canceled out, while important frequencies appear.” In the case of the Golden Gate Bridge, the researchers drove over the bridge 102 times using their automated devices, and the team used 72 trips made by Uber drivers with activated phones, the study said.
The team then compared the resulting data with that from a group of 240 sensors placed on the Golden Gate Bridge for three months.
The result, according to the study, was that the data from the phones merged with that from the bridge’s sensors; Of the 10 specific types of low-frequency vibrations that the engineers measured on the bridge, there was a close match, and in five cases, there was no difference between the methods at all.
“We were able to show that many of these times correspond very accurately to the bright waves of the bridge,” said Santi.
However, only 1 percent of all bridges in the US are suspension bridges. About 41 percent of the smallest bridges are concrete. So, the researchers also tested how well their method would work in that setting.
To do so, they tested a bridge in Ciampino, Italy, comparing the journeys of 280 vehicles on the bridge with six sensors placed on the bridge for seven months.
Again, the researchers were encouraged by these findings, although they found a 2.3 percent difference between the means of a specific modal frequency across all 280 courses, and a 5.5 percent difference over a smaller sample. That suggests that a larger volume of trips can yield more useful data.
“Our preliminary results suggest that only a small number of trips over a period of several weeks are sufficient to obtain useful information about the bridge’s mode frequencies,” Santi said.
If we look at the overall approach, Professor at MIT, Markus Buehler comments, “Dynamic signatures are emerging as a powerful tool for examining the properties of large and complex systems, from bacterial pathogens to the structural integrity of bridges as shown in this study. Said Buehler.
“It’s a universal signal that’s widely available in the natural and built environment that we’re just beginning to explore as a diagnostic and manufacturing tool in engineering,” Buehler said.
As Ratti admits, there are ways to refine and expand the research, including accounting for the effects of mounting a smartphone in a car, the effect of car type on the data, and more.
“We still have work to do, but we believe that our approach can be easily scaled up – up to the national level,” Ratti said.
“It may not reach the accuracy that one can use fixed sensors installed on the bridge, but it may be a very interesting early warning system. Small anomalies may suggest when an analysis should be done,” said Ratti.