Sensores ópticos baseados em LPFGs para identificação de condições de carregamento em estruturas

Abstract

To ensure the safety and durability of in service structures, constant monitoring of their structural health has been used by engineers to aid management, decision-making, and maintenance scheduling. In this context, sensing systems are extremely important. This work presents the use of long-period fiber grating (LPFGs) based optical fiber sensors on loading condition estimation. Optical fiber sensors are great candidates to real-time monitoring of structures because they are capable of handling harsh environments, have electromagnetic immunity, and low attenuation. This work shows two force identification scenarios: loading condition estimation at a simply-supported beam and two-dimensional compression force at an optical fiber. The first was accomplished by a mathematical model of the beam and was capable of estimating the force intensity with a relative mean absolute error of 1.22% and 2.51% for its position. Whereas the latter was accomplished by a machine learning data-driven model that estimated the force intensity with relative mean absolute error of 2.19% and 3.93% concerning its direction.