Adaptive sensor placement selection and response reconstruction for bridge structural health monitoring

Abstract Bridge structural health monitoring (SHM) measures the real-time responses of bridges via instrumented sensors. This paper focuses on the bridge displacement reconstruction at unmeasured locations of interest (LoIs) using measurements from sensors. Increasing the number of sensors on the bridge can capture more structural information. However, considering the complex structure and scale of bridges, minimizing the number of sensors and selecting the critical sensing locations (CSLs) to place sensors for measurements are significant in bridge SHM. To achieve efficient bridge SHM under limited number of sensors, this paper proposes an adaptive sensor selection and reconstruction neural network for determining the CSLs and reconstructing at all LoIs. In particular, a novel adaptive location selection mechanism as well as a unified network model and its learning process is proposed. In the experiments, the proposed model is validated based on a real-world case study of a long-span bridge in California, USA. Through the comparison of different sensor placement schemes. The experimental results show that the proposed model is effective and accurate in reconstructing the structural displacement responses via adaptive sensor selection for bridge SHM.