Static shape measurements using a multiplexed fiber Bragg grating sensor system

摘要

The research presented in this paper demonstrates the feasibility of detecting deformations of wing-like structures (cantilever plates) using a distributed, multiplexed fiber Bragg grating (FBG) sensor system. Overall, the work accomplished during the research effort demonstrates the ability of a multiplexed network of FBG sensors to measure strain across a structure and have those strains determine the deformation or shape of the structure. A neural network approach and a structural mechanics approach were examined to determine the deformation of the cantilever plate based on the recorded strain readings from the FBG sensors. This paper presents analytical and experimental results for both approaches. The experimental setup consisted of a cantilever plate with 36 FBG sensors mounted in longitudinal, transverse, and off-axis orientations on the plate's surface. The neural network method performed best when trained with experimentally measured strain and deflection data. The structural mechanics approach demonstrated more accurate comparisons to the ground truth shape, taken with a laser range finder, than the neural network approach. In all experimental cases using the structural mechanics approach, the root-mean-square error between the ground-truth measurements and the FBG sensor based measurements were less than 0.52 inches (approximately 10% of maximum deflection). In approximately 30% of the test cases, the root-mean-square error was less than 0.2 inches.