High Accuracy and Cost-Effective Fiber Optic Liquid Level Sensing System Based on Deep Neural Network
<p>Scheme diagram of the proposed multiple-point liquid level sensing system, which consists of four main blocks: (i) central office, (ii) sensing unit which shows the IDWM liquid level sensing system, (iii) preprocessing unit, and (iv) deep neural network structure.</p> "> Figure 2
<p>(<b>a</b>) The scheme of water level sensor structure that an FBG sensor connected to the indicated float. The water level can be controlled by the controlling gate. (<b>b</b>) The power versus wavelength of the reflected spectra of 10 different water level steps, and (<b>c</b>) the linear shift in wavelength when the water level change from 9.5 cm to 0.5 cm with the steps of 0.5 cm.</p> "> Figure 3
<p>The DNN model performance in terms of loss and accuracy for training and validation dataset.</p> "> Figure 4
<p>(<b>a</b>) The experimental result of the reflected spectrum of two water level sensors in five different water level steps of the first scenario. (<b>b</b>) The unmeasurable overlapping gap due to the overlap of two sensor spectra in the first scenario.</p> "> Figure 5
<p>(<b>a</b>) The experimental result of the reflected spectrum of two water level sensors in 5 different water level steps of the second scenario. (<b>b</b>) The unmeasurable overlapping gap due to the overlap of two sensor spectra in the second scenario.</p> "> Figure 6
<p>The predicted water level versus the actual water level for the (<b>a</b>) first scenario and (<b>b</b>) second scenario.</p> "> Figure 7
<p>The water level prediction error versus the water level for the (<b>a</b>) first scenario and (<b>b</b>) second scenario.</p> "> Figure 8
<p>Depiction of the mean absolute error (MAE) of experimental results for the support vector machine (SVM), random forest (RF), and deep neural network (DNN) algorithm.</p> "> Figure 9
<p>(<b>a</b>) The simulation results of reflected spectra of five water level sensors in five different water level steps, and (<b>b</b>) the predicted water level versus the actual water level of five sensors.</p> ">
Abstract
:1. Introduction
- To the best of our knowledge, we propose a liquid level sensing system for measuring the liquid level of multiple spots/containers at the same time for the first time;
- Since we propose a water level sensor that only consists of one FBG sensor for each spot, the sensing system is cost-effective;
- We propose DNN to predict the liquid level of each container in the liquid level sensing system, which is more accurate than conventional machine learning methods and our system’s accuracy is higher compared to earlier research that used AI to predict the water level.
2. Experimental Setup and Sensor Structure
3. DNN Model Configuration
4. Result and Discussion
4.1. First Scenario: When Both S11 and S21 Sensors’ Liquid Levels Change Simultaneously
4.2. Second Scenario: When Only the S11 Sensor Liquid Level Changes
4.3. Performance Evaluation
5. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Bao, L.; Dong, X.; Shum, P.P.; Shen, C. High Sensitivity Liquid Level Sensor Based on a Hollow Core Fiber Structure. Opt. Commun. 2021, 499, 127279. [Google Scholar] [CrossRef]
- Kumar, B.; Rajita, G.; Mandal, N. A Review on Capacitive-Type Sensor for Measurement of Height of Liquid Level. Meas. Control 2014, 47, 219–224. [Google Scholar] [CrossRef]
- Jin, B.; Liu, X.; Bai, Q.; Wang, D.; Wang, Y. Design and Implementation of an Intrinsically Safe Liquid-Level Sensor Using Coaxial Cable. Sensors 2015, 15, 12613–12634. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Woodard, S.E.; Taylor, B.D. A Wireless Fluid-Level Measurement Technique. Sens. Actuators A Phys. 2007, 137, 268–278. [Google Scholar] [CrossRef] [Green Version]
- Milesi, H.; Ventura, L.; Cavedo, F.; Norgia, M.; Dellaca, R.; Donati, S. Optical Flow Sensor for Lung Surfactant Delivery. In Proceedings of the 2019 IEEE International Conference on BioPhotonics, BioPhotonics 2019, Taipei, Taiwan, 15–18 September 2019; pp. 1–2. [Google Scholar]
- Panda, A.; Pukhrambam, P.D.; Keiser, G. Performance Analysis of Graphene-Based Surface Plasmon Resonance Biosensor for Blood Glucose and Gas Detection. Appl. Phys. A Mater. Sci. Process 2020, 126, 153. [Google Scholar] [CrossRef]
- Soundararajan, R.; Hsu, T.W.; Calderon-Delgado, M.; Donati, S.; Qin, Y.; Huang, S.L. Spectroscopic Full-Field Optical Coherence Tomography in Dermatology. In Proceedings of the 2019 IEEE International Conference on BioPhotonics, BioPhotonics 2019, Taipei, Taiwan, 15–18 September 2019; pp. 1–2. [Google Scholar]
- Nascimento, K.P.; Frizera-Neto, A.; Marques, C.; Leal-Junior, A.G. Machine Learning Techniques for Liquid Level Estimation Using FBG Temperature Sensor Array. Opt. Fiber. Technol. 2021, 65, 102612. [Google Scholar] [CrossRef]
- Sohn, K.R.; Shim, J.H. Liquid-Level Monitoring Sensor Systems Using Fiber Bragg Grating Embedded in Cantilever. Sens. Actuators A Phys. 2009, 152, 248–251. [Google Scholar] [CrossRef]
- Chang, H.Y.; Chang, Y.C.; Sheng, H.J.; Fu, M.Y.; Liu, W.F.; Kashyap, R. An Ultra-Sensitive Liquid-Level Indicator Based on an Etched Chirped-Fiber Bragg Grating. IEEE Photonics Technol. Lett. 2016, 28, 268–271. [Google Scholar] [CrossRef]
- Marques, C.A.F.; Pospori, A.; Sáez-Rodríguez, D.; Nielsen, K.; Bang, O.; Webb, D.J. Fiber Optic Liquid Level Monitoring System Using Microstructured Polymer Fiber Bragg Grating Array Sensors: Performance Analysis. In Proceedings of the 24th International Conference on Optical Fibre Sensors, SPIE, Curitiba, Brazil, 28 September–2 October 2015; Volume 9634, p. 96345V. [Google Scholar]
- He, R.; Teng, C.; Kumar, S.; Marques, C.; Min, R. Polymer Optical Fiber Liquid Level Sensor: A Review. IEEE Sens. J. 2022, 22, 1081–1091. [Google Scholar] [CrossRef]
- Lee, H.K.; Choo, J.; Shin, G.; Kim, J. Long-Reach DWDM-Passive Optical Fiber Sensor Network for Water Level Monitoring of Spent Fuel Pool in Nuclear Power Plant. Sensors 2020, 20, 4218. [Google Scholar] [CrossRef]
- Terzic, E.; Nagarajah, C.R.; Alamgir, M. Capacitive Sensor-Based Fluid Level Measurement in a Dynamic Environment Using Neural Network. Eng. Appl. Artif. Intell. 2010, 23, 614–619. [Google Scholar] [CrossRef]
- Pan, J.; Yin, Y.; Xiong, J.; Luo, W.; Gui, G.; Sari, H. Deep Learning-Based Unmanned Surveillance Systems for Observing Water Levels. IEEE Access 2018, 6, 73561–73571. [Google Scholar] [CrossRef]
- Qiao, G.; Yang, M.; Wang, H. A Water Level Measurement Approach Based on YOLOv5s. Sensors 2022, 22, 3714. [Google Scholar] [CrossRef] [PubMed]
- Yeh, C.H.; Guo, B.S.; Gu, C.S.; Chow, C.W.; Lin, W.P. Use of Same WDM Channels in Fiber Network for Bidirectional Free Space Optical Communication with Rayleigh Backscattering Interference Alleviation. IEEE Access 2019, 7, 169571–169576. [Google Scholar] [CrossRef]
- Keiser, G.E. A Review of WDM Technology and Applications. Opt. Fiber. Technol. 1999, 5, 3–39. [Google Scholar] [CrossRef] [Green Version]
- Chiu, P.H.; Lin, Y.S.; Manie, Y.C.; Li, J.W.; Lin, J.H.; Peng, P.C. Intensity and Wavelength-Division Multiplexing Fiber Sensor Interrogation Using a Combination of Autoencoder Pre-Trained Convolution Neural Network and Differential Evolution Algorithm. IEEE Photonics J. 2021, 13, 1–9. [Google Scholar] [CrossRef]
- Manie, Y.C.; Li, J.W.; Peng, P.C.; Shiu, R.K.; Chen, Y.Y.; Hsu, Y.T. Using a Machine Learning Algorithm Integrated with Data De-Noising Techniques to Optimize the Multipoint Sensor Network. Sensors 2020, 20, 1070. [Google Scholar] [CrossRef] [Green Version]
- Majumder, M.; Gangopadhyay, T.K.; Chakraborty, A.K.; Dasgupta, K.; Bhattacharya, D.K. Fibre Bragg Gratings in Structural Health Monitoring-Present Status and Applications. Sens. Actuators A Phys. 2008, 147, 150–164. [Google Scholar] [CrossRef]
- Li, B.; Tan, Z.-W.; Shum, P.P.; Wang, C.; Zheng, Y.; Wong, L. jie Dilated Convolutional Neural Networks for Fiber Bragg Grating Signal Demodulation. Opt. Express 2021, 29, 7110–7123. [Google Scholar] [CrossRef]
- Pak, M.; Kim, S. A Review of Deep Learning in Image Recognition. In Proceedings of the Proceedings of the 2017 4th International Conference on Computer Applications and Information Processing Technology, Kuta Bali, Indonesia, 8–10 August 2017; Volume 2018, pp. 1–3. [Google Scholar]
- Torkaman, P.; Yadav, G.S.; Wang, P.-C.; Lu, T.-Y.; Miao, X.-W.; Hsiao, F.-S.; Feng, K.-M.; Yang, S.-H. A 5G/Sub-Terahertz Heterogeneous Communication Network. IEEE Access 2022, 10, 65572–65584. [Google Scholar] [CrossRef]
- Liu, B.; Zeng, Q.; Lu, L.; Li, Y.; You, F. A Survey of Recommendation Systems Based on Deep Learning. IEEE Access 2021, 1754, 012148. [Google Scholar] [CrossRef]
- Lecun, Y.; Bengio, Y.; Hinton, G. Deep Learning. Nature 2015, 521, 436–444. [Google Scholar] [CrossRef] [PubMed]
- Alpaydin, E. Introduction to Machine Learning Ethem Alpaydin, 3rd ed.; MIT Press: Cambridge, MA, USA, 2014. [Google Scholar]
- Clevert, D.A.; Unterthiner, T.; Hochreiter, S. Fast and Accurate Deep Network Learning by Exponential Linear Units (Elus). arXiv 2015, arXiv:1511.07289. [Google Scholar] [CrossRef]
- Wang, Q.; Ma, Y.; Zhao, K.; Tian, Y. A Comprehensive Survey of Loss Functions in Machine Learning. Ann. Data Sci. 2022, 9, 187–212. [Google Scholar] [CrossRef]
- Murphy, K.P. Probabilistic Machine Learning: An Introduction; MIT Press: Cambridge, MA, USA, 2022. [Google Scholar]
DNN Parameters | Value |
---|---|
Number of neurons in the input layer | 1001 |
Number of hidden layers | 4 |
Number of neurons in each layer | 350 |
Activation function | ELU |
ELU hyperparameter (α) | 1 |
Optimizer algorithm | Adam |
Learning rate (η) | 0.001 |
Batch size | 250 |
Number of epochs | 500 |
Random Forest Parameters | Value |
---|---|
The number of trees in the forest | 502 |
The maximum depth of the tree | None |
The minimum number of samples required to split an internal node | 3 |
The minimum number of samples required to be at a leaf node | 1 |
SVM Parameters | Value |
---|---|
The kernel type | ‘rbf’ |
Epsilon | 0.0001 |
Kernel coefficient for ‘rbf’ (gamma) | 0.01 |
Tolerance for stopping criterion | 0.0001 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Dejband, E.; Manie, Y.C.; Deng, Y.-J.; Bitew, M.A.; Tan, T.-H.; Peng, P.-C. High Accuracy and Cost-Effective Fiber Optic Liquid Level Sensing System Based on Deep Neural Network. Sensors 2023, 23, 2360. https://doi.org/10.3390/s23042360
Dejband E, Manie YC, Deng Y-J, Bitew MA, Tan T-H, Peng P-C. High Accuracy and Cost-Effective Fiber Optic Liquid Level Sensing System Based on Deep Neural Network. Sensors. 2023; 23(4):2360. https://doi.org/10.3390/s23042360
Chicago/Turabian StyleDejband, Erfan, Yibeltal Chanie Manie, Yu-Jie Deng, Mekuanint Agegnehu Bitew, Tan-Hsu Tan, and Peng-Chun Peng. 2023. "High Accuracy and Cost-Effective Fiber Optic Liquid Level Sensing System Based on Deep Neural Network" Sensors 23, no. 4: 2360. https://doi.org/10.3390/s23042360