Presentation | 2021-11-19 Neural network-based strain measurement using Brillouin gain spectra calculated by non-uniform strain model Kurokawa Takaya, Narieda Shusuke, Naruse Hiroshi, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | The previously proposed methods of measuring temperature from the Brillouin gain spectrum (BGS) observed in an optical fiber using a neural network (NN) can be directly applied to the measurement of strain. Since the methods did not take temperature variation into account, the measurement error would be large when applied to BGS produced in a section where the strain varies, i.e., a non-uniform strain section. In this paper, we propose a new strain measurement method using the NN. A non-uniform strain model is constructed in which the strain is approximated by polygonal lines, and a mapping from the BGS calculated by the model to the strain is created by the NN. Numerical simulations are performed to investigate the effects of the noise of the BGS, the number of polygonal lines used for approximation, the change in strain, and the size of the observation region of the BGS on the strain measurement error. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Fiber optic sensing / Brillouin gain spectrum / Strain measurement / Polygonal line approximation / Model / Neural Network |
Paper # | OFT2021-52 |
Date of Issue | 2021-11-11 (OFT) |
Conference Information | |
Committee | OFT / OCS / IEE-CMN / ITE-BCT |
---|---|
Conference Date | 2021/11/18(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Online |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | Yuji Matsuura(Tohoku Univ.) / Takeshi Hoshida(Fujitsu) / Shun Morimura(CRIEPI) / 斎藤 恭一(NHK) |
Vice Chair | / / Kouji Hirata(Kansai Univ) |
Secretary | (NTT) / (Fujikura) / Kouji Hirata(NTT) / (Fujikura) |
Assistant | Yukihiro Goto(NTT) / Minoru Takahashi(Fujikura) / / Yuuichi Shinohara(TEPCO Power Grid) / Akihiro Tanaka(CRIEPI) / 濱住 啓之(東京高専) / 前田 直樹(住友電工) / 松本 卓三(古河電工) / 斉藤 一(テレビ東京) |
Paper Information | |
Registration To | Technical Committee on Optical Fiber Technology / Technical Committee on Optical Communication Systems / Technical Meeting on Communications / Technical Group on Broadcasting Technology |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Neural network-based strain measurement using Brillouin gain spectra calculated by non-uniform strain model |
Sub Title (in English) | |
Keyword(1) | Fiber optic sensing |
Keyword(2) | Brillouin gain spectrum |
Keyword(3) | Strain measurement |
Keyword(4) | Polygonal line approximation |
Keyword(5) | Model |
Keyword(6) | Neural Network |
1st Author's Name | Kurokawa Takaya |
1st Author's Affiliation | Mie University(Mie Univ) |
2nd Author's Name | Narieda Shusuke |
2nd Author's Affiliation | Mie University(Mie Univ) |
3rd Author's Name | Naruse Hiroshi |
3rd Author's Affiliation | Mie University(Mie Univ) |
Date | 2021-11-19 |
Paper # | OFT2021-52 |
Volume (vol) | vol.121 |
Number (no) | OFT-252 |
Page | pp.pp.7-12(OFT), |
#Pages | 6 |
Date of Issue | 2021-11-11 (OFT) |