Temperature Compensation for Optical Fiber Graphene Micro-Pressure Sensor Using Genetic Wavelet Neural Networks

Yixian Ge,Lingwen Shen,Mengmeng Sun

DOI: https://doi.org/10.1109/jsen.2021.3115810

IF: 4.3

2021-11-01

IEEE Sensors Journal

Abstract:Optical fiber sensors have numerous advantages and are widely used in several fields. A typical optic fiber Fabry-Perot (FP) sensor is used to determine the pressure and temperature. To improve the sensitivity and overcome various limitations of pressure- and temperature-sensitive sensors, in this study, we demonstrate a micro-pressure FP sensor fabricated on an optical fiber through a chemical etching process. A graphene diaphragm was used as a pressure-sensitive membrane. The influence of FP cavity's geometric parameters on the reflected signal was studied and simulated by following the optical transmission matrix theory. A finite element simulation of the model's deflection behavior was carried out through ANSYS static mechanics, which verified the pressure-sensitive model's accuracy. Experimental results show that the sensor exhibits high linearity and a sensitivity of 79.956 nm/kPa when the pressure ranges from 0 to 0.1 MPa. During pressure testing, a genetic algorithm-based wavelet neural network was used to compensate for temperature drifts in the optic fiber FP pressure sensors.