Radioelectronics and radioelectronic equipment manufacturing

Development of a Model Using a Multilayer Perceptron for Methane Concentration Measurement System Based on a Wireless Opto-Electronic Sensor

Authors

B. P. Knysh
Vinnytsia National Technical University ROR
Ya. A. Kulyk
Vinnytsia National Technical University ROR
DOI https://doi.org/10.31649/1997-9266-2025-182-5-192-199

Keywords

methane opto-electronic sensor concentration neural network multilayer perceptron

Abstract

The work proposes a model using a multilayer perceptron (MLP) for methane concentration measurement system based on a wireless optoelectronic sensor. In the controlled environment that may be dangerous to humans, a sensor is located, it remotely transmits information regarding the methane concentration to a computer, where a neural network processes the received data. As a sensor, it is proposed to use optoelectronic methane concentration sensor due to its high sensitivity, accuracy, reliability, and stability, as well as fast response, non-contact measurements, resistance to interference, the possibility of remote monitoring, and minimal maintenance. As a neural network, it is proposed to use a multilayer perceptron due to its simplicity in implementation and high efficiency for gas concentration estimation tasks in conditions of a small or medium training sample size, when the sum of basic nonlinearities well approximates the relationship between the sensor characteristics and the gas concentration without complex dynamic effects, and in case of low noise and interference, if the data has a moderate level of background fluctuations. Such systems allow measuring methane concentration with high accuracy by collecting data by means of optoelectronic sensor in real time, which, in combination with processing of the collected data using a neural network, demonstrates great potential in measuring methane concentration, and, consequently, early warning of its leakage. The work also proposes to use a multilayer perceptron model for processing data from a wireless optoelectronic sensor of methane concentration. The main indicators of the effectiveness of the proposed multilayer perceptron model, which were determined during the research, were selected such characteristics as the mean square error (MSE) and the mean absolute error (MAE). To analyze these indicators, the change in MSE and accuracy during model training and validation, the comparison of data from the optical sensor and the model, as well as the assessment of the change in MSE and MAE during the measurement process, were carried out.

200 77

How to Cite

[1]
“Development of a Model Using a Multilayer Perceptron for Methane Concentration Measurement System Based on a Wireless Opto-Electronic Sensor”, Вісник ВПІ, no. 5, pp. 192–199, Oct. 2025, doi: 10.31649/1997-9266-2025-182-5-192-199.

Author Biographies

B. P. Knysh, Vinnytsia National Technical University

Cand. Sc. (Eng.), Associate Professor of the Chair of General Physics

Ya. A. Kulyk, Vinnytsia National Technical University

Cand. Sc. (Eng.), Associate Professor of the Chair of Automation and Intelligent Information Technologies

References

Метан — отримання, властивості та застосування. [Електронний ресурс]. Режим доступу: https://www.products.pcc.eu/uk/blog/%d0%bc%d0%b5%d1%82%d0%b0%d0%bd-%d0%be%d1%82%d1%80%d0%b8%d0%bc%d0%b0%d0%bd%d0%bd%d1%8f-%d0%b2%d0%bb%d0%b0%d1%81%d1%82%d0%b8%d0%b2%d0%be%d1%81%d1%82%d1%96-%d1%82%d0%b0-%d0%b7%d0%b0%d1%81/ . Дата звернення 07.07.2025.
Г. Ю. Бахарєва, «Аналіз трьох основних джерел викидів метану як основа розробки технологій очистки повітря,» в Тезах доп. XXII міжнар. наук.-практ. конф., Харків, 15-17 жовт. 2014 р., у 4 ч. НТУ «ХПІ», № 18, 2014. http://www.kpi.kharkov.ua/archive/microcad/2014/S18/p250-p250.pdf .
K. Zhou, and Y. Liu, “Early-Stage Gas Identification Using Convolutional Long Short-Term Neural Network with Sensor Array Time Series Data,” Sensors, no. 21, рр. 1-17, 2021. https://doi.org/10.3390/s21144826 .
A.G. Howard, et al., “MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications,” ArXiv, no. 04, рр. 1-9, 2017. https://doi.org/10.48550/arXiv.1704.04861 .
X. Zhang, X. Zhou, M. Lin, and J. Sun, “ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018, рр. 18-23. https://doi.org/10.48550/arXiv.1707.01083 .
M. Zhou, S. Wang, J. Li, Z. Wei, and L. Shui, “A Wireless Sensor Network-Based Combustible Gas Detection System Using PSO-DBO-Optimized BP Neural Network,” Sensors, no. 10, рр. 1-23, 2025. https://doi.org/10.3390/s25103151 .
S. Song, et al., “Research on a working face gas concentration prediction model based on LASSO-RNN time series data,” Cell Press, no. 9, рр. 1-10, 2023. https://doi.org/10.1016/j.heliyon.2023.e14864 .
K. Gao, Z. Zhou, and Ya. Qin, “Gas concentration prediction by LSTM network combined with wavelet thresholding denoising and phase space reconstruction,” Cell Press, no. 10, рр. 1-10, 2024. https://doi.org/10.1016/j.heliyon.2024.e28112 .
Б. Р. Водяник, М. Г. Лорія, і Є. В. Кобзарев, «Застосування фізично обґрунтованих нейронних мереж (PINN) у контролі якості продукту процесу Габера-Боша,» Вісник Східноукраїнського університету імені Володимира Даля, № 6, с. 190-198, 2024. https://doi.org/10.33216/1998-7927-2024-286-6-190-198 .
S. Hoffman, and R. Jasinski, “The Use of Multilayer Perceptrons to Model PM2.5 Concentrations at Air Monitoring Stations in Poland,” Atmosphere, no. 14, рр. 1-19, 2023. https://doi.org/10.3390/s25103151 .
J. D. Kim, H. G. Byun, and K. C. Persaud, “Application of a multilayer perceptron based on the Levenberg-Marquardt algorithm to odour pattern classification and concentration estimation using odour sensing system,” in 7th International Symposium on Olfaction and Electronic Noses, 2000. [Electronic resource]. Available: https://research.manchester.ac.uk/en/publications/application-of-a-multilayer-perceptron-based-on-the-levenberg-mar .
M. Tutak, T. Krenicky, R. Pirnik, J. Brodny, and W. Grebski, “Predicting Methane Concentrations in Underground Coal Mining Using a Multi-Layer Perceptron Neural Network Based on Mine Gas Monitoring Data,” Sustainability, no. 16, рр. 1-23, 2024. https://doi.org/10.3390/su16198388 .
T. Jia, et al., “Mixed Natural Gas Online Recognition Device Based on a Neural Network Algorithm Implemented by an FPGA,” Sensors, no. 19, рр. 1-11, 2019. https://doi.org/10.3390/s19092090 .
Q. Chi, et al., “Dual-Comb Gas Sensor Integrated with a Neural Network-Based Spectral Decoupling Algorithm of Overlapped Spectra for Gas Mixture Sensing,” ACS OMEGA, no. 8, рр. 14648-14655, 2023. https://doi.org/10.1021/acsomega.3c00518 .
Б. П. Книш, Й. Й. Білинський, і В. Р. Гладишевський, «Оптичний сенсор концентрації газу», Патент України G01N 21/81. № 61667МПК (2011), 25.07.2011.
Й. Й. Білинський, Б. П. Книш, і О. А. Павлюк, «Інфрачервоний двохвильовий сенсор контролю концентрації газу,» Наукові праці Вінницького національного технічного університету, № 3, 2011. [Електронний ресурс]. Режим доступу: http://praci.vntu.edu.ua/index.php/praci/article/view/290 .
ДСТУ EN 16723-1:2023. «Природний газ і біометан для використання в транспорті та біометан для закачування в мережу природного газу. Частина 1. Технічні характеристики біометану для закачування в мережу природного газу.» Чинний від 2023-11-01. Вид. офіц. Київ : УкрНДНЦ, 2023. 32 с.

Similar Articles

1-10 of 50

You may also start an advanced similarity search for this article.

Most read articles by the same author(s)

1 2 > >>