Design of a hetero-core smart fiber optic microbend sensor

Abstract

In this work, hetero-core optical fibers were used instead of multimode optical fibers in microbend sensors and it was concluded from the experiments that, using hetero-core fiber instead of multimode fiber with the same core diameter for microbending sensors, turns out to be a more sensitive sensor configuration. Multimode fibers were used instead of single mode fibers in the hetero-core section of the hetero-core optical fibers for the first time. In addition to that, by using the data taken from microbend force sensors, the ability of Artificial Neural Networks (ANNs) in the prediction of force values was also analyzed. Multilayer Perceptron with different algorithms were used as a ANN model. All algorithms used can predict force values with considerable errors. Furthermore, an intelligent and multiple sensor architecture were suggested by using ANNs to get more sensitive and accurate measurement results.