A new system to assess the structural health of bridges using Artificial Intelligence (AI) and an array of sensors is being developed by a team at the University of Texas at Arlington.
Modern bridges are typically built with weight-in-motion systems including sensors that measure vibrations, strain and deflection.
By measuring the bridge’s response to these elements, they can estimate the gross vehicle weights of passing vehicles and their effects on a bridge’s structural health.
But the researchers said these sensors don’t take into account the different types of trucks, multiple lanes, times of day and how heavy traffic is.
Nevertheless, since these sensors are often already in place, the team are trying to create a system by which these traditional measurements of structural health can be refined through machine learning.
With the resulting data, transportation departments could set more accurate load parameters for bridges and get a better picture of a structure’s overall integrity.
Such data could prevent disasters like the 2018 collapse of the Ponte Morandi bridge (pictured) in Genoa, Italy when its weakening supports were put under additional pressure during a heavy storm.
“We are combining a physics-based model with artificial intelligence, because the more a computer learns, the better information you get,” lead researcher Suyun Ham said. “Ultimately, the addition of machine learning allows us to accurately determine multiple conditions.”
He is also working on a non-contact testing system to make faster, easier and more accurate determinations about when and where bridge repairs are needed.