Bridge load rating using dynamic response collected through wireless sensor networks

Abstract

This paper describes a method for load rating of prestressed box beam (PSBB) bridges based on their dynamic response collected using wireless sensor networks (WSNs). Although the percentage of deficient bridges in the United States has been declining slowly, a significant portion is still closed to traffic or posted with load restrictions. An accurate load rating of bridges is very expensive; therefore, new tools for quick, efficient and response-based load rating of bridges will save time and money. The hypothesis is based on the assumption that the health of a bridge is associated with its vibration signatures under vehicular loads. Two WSNs were deployed on a 25-year old PSBB bridge under trucks with variable loads and speeds for collecting real-time dynamic response at the current condition. Dynamic simulations of three dimensional finite element models of a bridge were performed to acquire its dynamic response under vehicular loads at its newest condition right after construction. The bridge model was validated by field testing and numerical analysis. Fast Fourier Transform and peak-picking algorithms were used to find maximum peak amplitudes and their corresponding frequencies. This information and the necessary bridge geometric parameters were used to calculate the in-service stiffness of the bridge in order to develop application software for load rating of bridges. The application software can instantly calculate the load rating of a PSBB bridge by collecting its real time dynamic response under vehicular loads using WSNs. The research outcome and the software will help reduce bridge maintenance costs and will increase public safety.