Laser & Optoelectronics Progress, Volume. 60, Issue 7, 0706009(2023)

Ultrasensitive Methane Volume Fraction Sensor Based on Long Period Fiber Grating and Back-Propagation Neural Network

Chao Du1、*, Bin Zhang2, Shuang Zhao1, Qiuyu Wang1, Li Zhang1, Liqin Cui1, and Xiao Deng1、**
Author Affiliations
  • 1College of Physics and Optoelectronics, Taiyuan University of Technology, Taiyuan 030024, Shanxi, China
  • 2Military Representative Office of PLA Air Force Equipment Department in Taiyuan, Taiyuan 030006, Shanxi, China
  • show less

    An ultrasensitive long period fiber grating (LPFG) sensor is proposed and investigated for the measurement of methane volume fraction by applying the selective adsorption property of cryptophane-E to methane. The cladding diameter is reduced to make the low order cladding mode LP06 work near dispersion turning point (DTP). The TiO2 thin film with optimized thickness is coated on the surface to ensure coupled cladding mode LP06 within the mode transition (MT) region, which can result in higher refractive index (RI) sensitivity of LPFG. The methane gas changes the RI of the cryptophane-E that is coated on the surface of the LPFG sensor, and then the methane volume fraction can be measured by monitoring the shift of the resonance wavelength. A high sensitivity of 249.6 nm/% can be achieved when the methane volume fraction changes from 0% to 3.5% owing to the contribution of DTP and MT. A back-propagation (BP) neural network is designed for the nonlinearity response of the sensor at different volume fraction. The result shows that a maximum predicted error of 0.008% is recorded in the methane volume fraction change range. The excellent performance shows that the proposed sensor has potential application value in the field of coal mine safety monitoring.

    Tools

    Get Citation

    Copy Citation Text

    Chao Du, Bin Zhang, Shuang Zhao, Qiuyu Wang, Li Zhang, Liqin Cui, Xiao Deng. Ultrasensitive Methane Volume Fraction Sensor Based on Long Period Fiber Grating and Back-Propagation Neural Network[J]. Laser & Optoelectronics Progress, 2023, 60(7): 0706009

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Fiber Optics and Optical Communications

    Received: Oct. 10, 2022

    Accepted: Nov. 30, 2022

    Published Online: Mar. 31, 2023

    The Author Email: Du Chao (duchao@tyut.edu.cn), Deng Xiao (dengxiao@tyut.edu.cn)

    DOI:10.3788/LOP222732

    Topics