An optimal sensor design framework accounting for sensor reliability over the structural life cycle

Abstract

One of the most prominent goals of a structural health monitoring (SHM) system is to infer the state of the structure to inform appropriate maintenance actions that affect performance and safety over the life cycle of the structure. SHM necessarily involves data acquisition at the beginning of the SHM workflow, but damage state inference may be fundamentally flawed if the sensing system itself initiates unreliable data. The operational and environmental conditions that these sensors face, in addition to normal manufacturing defects, may result in varying functionality in both space and time. Therefore, it becomes imperative to account for sensor reliability in the optimal sensor design process for the SHM system at the outset – that is, consider the possibility of sensors malfunctioning in the initial/pre-posterior design stage without the benefit of previous deployment information – such that the initial design is robust to sensor malfunction with regard to decision performance objectives. This paper details an optimal sensor design framework at the pre-posterior stage that incorporates the time-dependent reliability of the sensor network over the life cycle of the structure. The proposed framework can be used to obtain an intelligent initial design that can be updated with time, based on the principles and procedure presented here, but by considering the updated/current state of the structure. The generic framework is laid out and is finally demonstrated on a complex real-world miter-gate structure.