Accurate Strain Sensing With Fiber-Optic Fabry-Perot Sensors Based on CNN-LSTM Model

被引:8
作者
Wei, Chuanhao [1 ]
Liu, Qiang [1 ]
Wang, Yiping [1 ]
Zhu, Dan [1 ]
Shi, Jingzhan [1 ]
Lin, Dongdong [1 ]
机构
[1] Nanjing Normal Univ, Sch Comp & Elect Informat Artificial Intelligence, Nanjing 210023, Peoples R China
基金
中国国家自然科学基金;
关键词
Sensors; Optical fiber sensors; Optical fibers; Demodulation; Strain; Optical interferometry; Optical fiber networks; Convolutional neural network (CNN)-long short-term memory (LSTM) model; fiber-optics Fabry--Perot (FP) sensor; strain extraction; INTERFEROMETER;
D O I
10.1109/JSEN.2024.3386724
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
In this work, we proposed a new method of demodulating the spectrum of fiber-optic Fabry-Perot (FP) strain sensors by employing convolutional neural networks (CNNs) adding long short-term memory (LSTM) networks, hereinafter referred to as CNN-LSTM. The model was built to retrieve the spectra of different strains since the peak tracking algorithm may difficult to accurately find the peaks of spectra and some traditional algorithms have their own limitations, which may cause large errors in demodulation results. The spectra of strain will be normalized first and then sent to the model for training, the strain information can be directly extracted without any peak wavelength or cavity length as an intermediary quantity, which simplifies the demodulation process. Experimental results show that the accuracy represented by the coefficient of determination (R-2) better than 99.99%, and the root mean square error (RMSE) is about 0.524 mu epsilon. Additionally, the model has comprehensive advantages over other models in terms of training cost and accuracy. The proposed model demonstrates good performance based on low-resolution spectra, showing its great potential in developing low-cost sensing systems.
引用
收藏
页码:17725 / 17732
页数:8
相关论文
共 36 条
[21]  
Lu H. S., 2007, Research on Multipoint Measurement Technology of Fiber Optic FARPER Strain Sensing
[22]   Deep Learning Assisted Pressure Sensing Based on Sagnac Interferometry Realized by Side-Hole Fiber [J].
Mei, Yongchang ;
Zhang, Shengqi ;
Cao, Zihan ;
Xia, Titi ;
Yi, Xingwen ;
Liu, Zhengyong .
JOURNAL OF LIGHTWAVE TECHNOLOGY, 2023, 41 (02) :784-793
[23]   High Accuracy Pressure Sensing With Sagnac Interferometry Based On Deep Learning Approach [J].
Mei, Yongchang ;
Zhang, Shengqi ;
Cao, Zihan ;
Xia, Titi ;
Yi, Xingwen ;
Liu, Zhengyong .
2022 IEEE 24TH INTERNATIONAL WORKSHOP ON MULTIMEDIA SIGNAL PROCESSING (MMSP), 2022,
[24]   Laser-micromachined Fabry-Perot optical fiber tip sensor for high-resolution temperature-independent measurement of refractive index [J].
Ran, Z. L. ;
Rao, Y. J. ;
Liu, W. J. ;
Liao, X. ;
Chiang, K. S. .
OPTICS EXPRESS, 2008, 16 (03) :2252-2263
[25]  
Song N., 2021, Research on Measurement Method of Fiber Optic Fabry-Perot (FP) Pressure Sensor Based on Machine Learning Algorithm
[26]   Fiber Sagnac interferometer temperature sensor [J].
Starodumov, AN ;
Zenteno, LA ;
Monzon, D ;
DeLaRosa, E .
APPLIED PHYSICS LETTERS, 1997, 70 (01) :19-21
[27]   Recent Advances in Machine Learning for Fiber Optic Sensor Applications [J].
Venketeswaran, Abhishek ;
Lalam, Nageswara ;
Wuenschell, Jeffrey ;
Ohodnicki, P. R., Jr. ;
Badar, Mudabbir ;
Chen, Kevin P. ;
Lu, Ping ;
Duan, Yuhua ;
Chorpening, Benjamin ;
Buric, Michael .
ADVANCED INTELLIGENT SYSTEMS, 2022, 4 (01)
[28]   Speaker Recognition of Fiber-Optic External Fabry-Perot Interferometric Microphone Based on Deep Learning [J].
Wang, Yangfeng ;
Wan, Shengpeng ;
Zhang, Sijun ;
Yu, Junsong .
IEEE SENSORS JOURNAL, 2022, 22 (13) :12906-12912
[29]  
Wenjie L., 2022, Research on the Demodulation System of Groundwater Temperature and Pressure Dula-detection Fiber-Optic Sensor
[30]   An arrayed waveguide grating based multiplexer and interrogator for Fabry-Perot sensors [J].
Willshire, AJ ;
Niewczas, P ;
McDonald, JR .
IEEE SENSORS JOURNAL, 2005, 5 (05) :964-969