基于改进LSTM的FBG传感网络光谱基线校正方法

韩颖; 张旭; 于明鑫; 庄炜

doi:10.13756/j.gtxyj.2024.230032

您当前的位置：

首页 >

文章列表页 >

基于改进LSTM的FBG传感网络光谱基线校正方法

光通信系统与网络技术 | 更新时间：2024-08-05

- 基于改进LSTM的FBG传感网络光谱基线校正方法
- Baseline Correction Method of FBG Sensor Network Spectrum based on the Improved LSTM Model
- 光通信研究 2024年第4期页码：23003201-23003208
- 作者机构：
  
  1.北京信息科技大学　光电测试技术与仪器教育部重点实验室，北京　100192
  2.北京信息科技大学　光纤传感与系统北京实验室，北京　100016
  3.天津大学，天津　300072
- 作者简介：
  
  韩颖（1993－），男，广东深圳人。硕士，主要研究方向为光纤传感与光电器件。
  张旭，博士。E-mail：zhangxubistu@outlook.com
- 基金信息：
  
  北京市自然基金－市教委联合基金资助项目(KZ201911232044)
- DOI：10.13756/j.gtxyj.2024.230032
  中图分类号： TN929
- 纸质出版日期：2024-08-10，
  
  收稿日期：2023-05-01，
  
  修回日期：2023-05-16，
- 稿件说明：
扫描看全文
韩颖，张旭，于明鑫，等. 基于改进LSTM的FBG传感网络光谱基线校正方法[J]. 光通信研究，2024(4)：230032.

Han Y, Zhang X, Yu M X, et al. Baseline Correction Method of FBG Sensor Network Spectrum based on the Improved LSTM Model [J]. Study on Optical Communications, 2024(4)：230032.
韩颖，张旭，于明鑫，等. 基于改进LSTM的FBG传感网络光谱基线校正方法[J]. 光通信研究，2024(4)：230032. DOI： 10.13756/j.gtxyj.2024.230032.

Han Y, Zhang X, Yu M X, et al. Baseline Correction Method of FBG Sensor Network Spectrum based on the Improved LSTM Model [J]. Study on Optical Communications, 2024(4)：230032. DOI： 10.13756/j.gtxyj.2024.230032.

摘要

【目的】

针对复杂外界环境所致的光纤布拉格光栅（FBG）光谱信号基线漂移问题，文章提出了一种基于改进的长短期记忆（LSTM）模型的光谱基线校正方法。

【方法】

改进的LSTM模型由卷积神经网络（CNN）、全连接层和LSTM网络组成。CNN与LSTM模型相比，改进的LSTM模型利用CNN提取FBG光谱信号的特征信息。文章使用仿真数据集和实测数据集对改进的LSTM模型进行训练。仿真数据集由特征噪声、基线和FBG光谱组成。分别使用小波软阈值法、惩罚最小二乘法、循环神经网络（RNN）法、LSTM法和改进的LSTM法进行基线校正。使用FBG传感信号存活率和均方根误差（RMSE）对5种方法的校正结果进行评价。

【结果】

使用仿真数据集对LSTM模型进行训练与测试，改进的LSTM模型将FBG传感信号存活率提高了60.8%。采用仿真数据集对模型进行预训练后，再将使用实测数据集对经过预训练的模型进行训练得到的改进LSTM模型与直接使用实测数据集训练的模型相比，改进的LSTM模型具有更好的校正效果。FBG光谱的RMSE降低了10.95%，RMSE的标准差降低了4%。使用改进的LSTM模型对FBG光谱实测数据集进行校正，FBG传感信号存活率提高了50.5%。与小波软阈值法、惩罚最小二乘法、RNN法和LSTM法相比，改进后的LSTM模型具有更好的校正效果，RMSE的均值和标准差分别为0.012 2和0.002 4。解调中心波长的RMSE为0.036 pm，并且基线校正过程只需9.68 ms。

【结论】

改进的LSTM模型是一种有效的基线校正方法，在复杂外部环境下具有广阔的应用前景。

Abstract

【Objective】

The baseline drift of the Fiber Bragg Grating (FBG) spectral signal is usually one of the main problems

caused by the complex external environment. A spectral baseline correction method based on the improved Long Short Term Memory (LSTM) model is proposed in this paper.

【Methods】

Compared with LSTM model

the improved LSTM model extracts feature information of FBG spectral signal by the Convolutional Neural Network (CNN). The improved LSTM model is composed of CNN

full connection

and LSTM network. In this paper

the improved LSTM model is trained by artificial datasets and measured datasets. The artificial datasets are made up of feature noise

baseline

and FBG spectroscopy. Five methods including wavelet soft threshold method

penalty least square method

Recurrent Neural Network (RNN)

LSTM

and the improved LSTM model are used as baseline correction. Identification signal probability and Root Mean Square Error (RMSE) are used to evaluate correction results by the five methods.

【Results】

The artificial datasets of FBG signal are corrected by the improved LSTM model

and the identification signal probability is increased by 60.8%. The improved LSTM model with training by artificial datasets and measured datasets shows better correction results

compared with training by measured datasets. The mean of the RMSE for FBG spectrum decreases by 10.95%. The standard deviation of RMSE decreases by 4%. The measured datasets of FBG signal are corrected by the improved LSTM model

and the identification signal probability is increased by 50.5%. Compared with wavelet soft threshold method

penalty least square method

RNN and LSTM

the improved LSTM model shows best correction results. The mean values of RMSE and the standard deviation of RMSE are 0.012 2 and 0.002 4

respectively. The RMSE value of the demodulated central wavelength is 0.036 pm. And the baseline correction process takes only 9.68 ms.

【Conclusion】

The improved LSTM model is an effective method to achieve baseline correction

and has wide range of application prospects in complex external environment.

关键词

光纤布拉格光栅光谱基线校正改进长短期记忆模型深度学习

Keywords

FBGspectral baseline correctionimproved LSTM modelsdeep learning

references

刘铁根, 王双, 江俊峰, 等. 航空航天光纤传感技术研究进展[J]. 仪器仪表学报, 2014, 35(8):1681-1692.

Liu T G, Wang S, Jiang J F, et al. Advances in Optical Fiber Sensing Technology for Aviation and Aerospace Application[J]. Chinese Journal of Scientific Instrument, 2014, 35(8):1681-1692.

曲道明, 孙广开, 李红, 等. 变形机翼柔性蒙皮形状光纤传感及重构方法[J]. 仪器仪表学报, 2018, 39(1):144-151.

Qu D M, Sun G K, Li H, et al. Optical Fiber Sensing and Reconstruction Method for Morphing Wing Flexible Skin Shape [J]. Chinese Journal of Scientific Instrument, 2018, 39(1):144-151.

田石柱, 邱伟宸, 温科, 等. FBG传感器关于裂缝及损伤的监测应用研究[J]. 激光技术, 2017, 41(1):129-132.

Tian S Z, Qiu W C, Wen K, et al. Research on the Application of FBG Sensor on Crack and Damage Monitoring [J]. Laser Technology, 2017, 41(1):129-132.

王瑨, 王永杰, 张登攀. 面向海洋应用的光纤光栅温度传感器研究进展[J]. 激光与红外, 2019, 49(5):515-522.

Wang J, Wang Y J, Zhang D P. Research Progress of Fiber Grating Temperature Sensor for Marine Applications [J]. Laser & Infrared, 2019, 49(5):515-522.

李延河, 杨战标, 朱元广, 等. 基于弱光纤光栅传感技术的围岩变形监测研究[J]. 煤炭科学技术, 2023, 51(6):11-19.

Li Y H, Yang Z B, Zhu Y G, et al. Research on Deformation Monitoring of Surrounding Rock based on Weak Fiber Grating Sensing Technology [J]. Coal Science and Technology, 2023, 51(6):11-19.

Meng S J, Sun Y Q, Wang M. Fiber Bragg Grating Sensors for Subgrade Deformation Monitoring in Seasonally Frozen Regions[J]. Structural Control and Health Monitoring, 2020, 27(1):e2472.

Tosi D, Schena E, Molardi C, et al. Fiber Optic Sensors for Sub-centimeter Spatially Resolved Measurements: Review and Biomedical Applications [J]. Optical Fiber Technology, 2018, 43:6-19.

秦根朝, 孟凡勇, 李红, 等. 光纤光栅链路反射谱强度自适应解调[J]. 红外与激光工程, 2022, 51(5):245-252.

Qin G C, Meng F Y, Li H, et al. Adaptive Demodulation of FBG Link Reflectance Spectrum Intensity [J]. Infrared and Laser Engineering, 2022, 51(5):245-252.

楚奇梁, 刘琨, 江俊峰, 等. 基于可调谐激光器的光纤光栅波长解调系统误差分析与补偿[J]. 光电子·激光, 2018, 29(1):14-21.

Chu Q L, Liu K, Jiang J F, et al. Error Analysis and Compensation of Fiber Grating Wavelength Demodulation System based on Tunable Laser [J]. Optoelectronics & Lasers, 2018, 29(1):14-21.

方勇华, 孔超, 兰天鸽, 等. 应用小波变换实现光谱的噪声去除和基线校正[J]. 光学精密工程, 2006, 14(6):1088-1092.

Fang Y H, Kong C, Lan T G, et al. Denoising and Baseline Correction of Spectrum by Wavelet Transform [J]. Optics and Precision Engineering, 2006, 14(6):1088-1092.

赵恒, 陈娱欣, 续小丁, 等. 基于局部对称重加权惩罚最小二乘的拉曼基线校正[J]. 中国激光, 2018(12):280-291.

Zhao H, Chen Y X, Xu X D, et al. Raman Baseline Correction based on Local Weighing Weighted Penalty Least Squares [J]. Chinese Journal of Lasers, 2018(12):280-291.

吕峥, 庄炜, 吴越, 等. 基于改进EMD方法的FBG传感网络光谱基线校正研究[J]. 仪器仪表学报, 2022, 43(1):190-197.

Lü Z, Zhuang W, Wu Y, et al. Research on Spectral Baseline Correction of FBG Sensor Network based on Improved EMD Method [J]. Chinese Journal of Scientific Instrument, 2022, 43(1):190-197.

潘成胜, 王羽夫, 杨力. 基于改进LSTM算法的天地一体化信息网络流量预测[J]. 天地一体化信息网络, 2020, 1(2):57-65.

Pan C S, Wang Y F, Yang L. Traffic Prediction of Space-ground Integrated Information Network based on Improved LSTM Algorithm [J]. Space-Integrated-Ground Information Network, 2020, 1(2):57-65.

于海英, 王文斌, 解全才, 等. 基于长短期记忆模型LSTM的近断层强震动记录零基线校正方法[J]. 地震工程与工程振动, 2022, 42(4):35-42.

Yu H Y, Wang W B, Xie Q C, et al. Zero Baseline Correction Method for Near-fault Strong Seismic Recording based on Long Short-term Memory Model LSTM [J]. Earthquake Engineering and Engineering Vibration, 2022, 42(4):35-42.

范竣翔, 李琦, 朱亚杰, 等. 基于RNN的空气污染时空预报模型研究[J]. 测绘科学, 2017(7):76-83.

Fan J X, Li Q, Zhu Y J, et al. Research on Spatiotemporal Prediction Model of Air Pollution based on RNN [J]. Science of Surveying and Mapping, 2017(7):76-83.

武丽芬, 严学勇, 赵吉. 基于RNN模型与LSTM模型的机器作诗研究[J]. 科技创新与应用, 2021, 11(27):48-50.

Wu L F, Yan X Y, Zhao J. Research on Machine Poetry based on RNN Model and LSTM Model [J]. Science and Technology Innovation and Application, 2021, 11(27):48-50.

姜建国, 万成德, 陈鹏, 等. 基于1D CNN-LSTM和迁移学习的短期电力负荷预测[J]. 吉林大学学报(信息科学版), 2023, 41(1):124-130.

Jiang J G, Wang C D, Chen P, et al. Short-Term Power Load Prediction based on 1D CNN-LSTM and Transfer Learning [J]. Journal of Jilin University (Information Science Edition), 2023, 41(1):124-130.

叶壮, 余建波. 基于多通道一维卷积神经网络特征学习的齿轮箱故障诊断方法[J]. 振动与冲击, 2020, 39(20):55-66.

Ye Z, Yü J B. Gearbox Fault Diagnosis Method based on Multi-channel One-dimensional Convolutional Neural Network Feature Learning [J]. Journal of Vibration and Shock, 2020, 39(20):55-66.

陈伟, 陈克松, 纪青春, 等. 基于1D-CNN＋GRU的光伏阵列故障诊断方法研究[J]. 自动化仪表, 2022, 43(6):13-17.

Chen W, Chen K S, Ji Q C, et al. Research on Fault Diagnosis Method of Photovoltaic Array based on 1D-CNN＋GRU [J]. Automation Instrument, 2022, 43(6):13-17.

车畅畅, 王华伟, 倪晓梅, 等. 基于1D-CNN和Bi-LSTM的航空发动机剩余寿命预测[J]. 机械工程学报, 2021, 57(14):304-312.

Che C C, Wang H W, Ni X M, et al. Remaining Life Prediction of Aero-engine based on 1D-CNN and Bi-LSTM[J]. Journal of Mechanical Engineering, 2021, 57(14):304-312.

浏览量

下载量

CSCD

CNKI被引量

文章被引用时，请邮件提醒。

提交

工具集

关联资源

面向光纤光栅传感线性拟合度的PSO-GPR算法

光纤布拉格光栅压力传感技术与应用进展

电力线载波通信中基于深度学习的信道估计

深度学习在光纤通信系统损伤补偿中的应用

基于深度学习的电光调制型可编程光学频率梳