A two-stage optimal sensor placement method for multi-type structural response reconstruction

Optimal multi-type sensor placement has gained considerable attention in the structural health monitoring field during the past few years. Although a structural response reconstruction-oriented optimization method for a multi-type sensor has been developed, the challenge of information redundancy in the collected signal deserves further investigation. To tackle this challenge, this paper presents a two-stage optimization framework for response reconstruction with the capability to reduce the multi-type sensor information redundancy. In the first stage, the optimization of multi-type sensor placement for response reconstruction is performed to initially determine the optimal sensor deployment scheme. After the optimal sensor locations are selected, the second-stage optimization introduces a metric, called the distance coefficient, to evaluate the information independence level between sensor locations with the goal of reducing the sensor information redundancy. A numerical study on a bridge model is first performed to evaluate the feasibility of the proposed framework, after which a lab-scale physical bridge model is tested to validate its effectiveness. Both the numerical and experimental results demonstrate that the proposed two-stage optimization framework can reduce the sensor information redundancy and, at the same time, produce a satisfying result in the response reconstruction of key locations.