«Automation and Informatization of the fuel and energy complex»

A model for determining readiness class of thermochemical gas detector sensors using a convolutional neural network with the input set packet splitting into packets for reducing zero drift effect

UDC: 681.5:519.86
DOI: -

Authors:

¹ National University of Oil and Gas "Gubkin University", Moscow, Russia

Keywords: automated process control systems, convolutional neural networks (CNN), thermochemical sensors (TCS), zero drifting, quality of readiness class forecasting

Annotation:

The author of the article describes an approach and a mathematical model that contribute to improving the quality of convolutional neural networks (CNN) used to predict the states of thermochemical sensors (TCS), which determine the class of its readiness. TCS are widely used at the facilities of the fuel and energy complex (FEC) to analyze the environment and identify pre-explosive concentrations at the first level of informing automated process control systems. They are an important part of ensuring fire safety at fuel and energy complex facilities. The main problem of determining their readiness was and still remains the so-called zero drift – a change of the their sensitive element (SE) state under the influence of various environmental factors or maintenance measures. Smell detection devices (electronic noses) also have a similar disadvantage. The readiness class of a TCS is considered to be such values of its parameters that set its service life before replacement, taking into account the wear of the SE. Since predicting real states of CNN requires a large number of features for classification, different forms of itsr training may not always give the required results. The forms of CNN models proposed in the article are designed to improve the quality of forecasting TCS readiness class.

Bibliography:

1. Abrosimov A.A., Topol'skiy N.G., Fedorov A.V. Avtomatizirovannye sistemy pozharovzryvobezopasnosti neftepererabatyvayushchikh proizvodstv. – M.: Akademiya GPS MVD Rossii, 2000. – 239 s.
2. Samarin I.V. Metody, modeli i algoritmy avtomatizatsii organizatsionnogo upravleniya pozharovzryvobezopasnost'yu ob"ektov toplivno-energeticheskogo kompleksa: dis. … d-ra tekhn. nauk: 2.3.4; 2.3.3. – M., 2022. – 446 s. – URL: https://www.dissercat.com/content/metody-modeli-i-algoritmy-avtomatizatsii-organizatsionnogo-upravleniya-pozharovzryvobezopasn (data obrashcheniya 01.08.2025).
3. GOST 12.1.044-89 (ISO 4589-84). Sistema standartov bezopasnosti truda. Pozharovzryvoopasnost' veshchestv i materialov. Nomenklatura pokazateley i metody ikh opredeleniya. – Vved. 1991–01–01. – M.: Standartinform, 2006. – 99 s.
4. GOST R 52350.29.2-2010. Vzryvoopasnye sredy. Chast' 29-2. Gazoanalizatory. Trebovaniya k vyboru, montazhu, primeneniyu i tekhnicheskomu obsluzhivaniyu gazoanalizatorov goryuchikh gazov i kisloroda. – Vved. 2012–01–01. – M.: Standartinform, 2011. – VI, 75 s.
5. Vedomstvennye normy VSN 64-86. Metodicheskie ukazaniya po ustanovke signalizatorov i gazoanalizatorov kontrolya dovzryvoopasnykh i predel'no dopustimykh kontsentratsiy khimicheskikh veshchestv v vozdukhe proizvodstvennykh pomeshcheniy. – Vved. 1986–07–01. – URL: https://files.stroyinf.ru/Data2/1/4293823/4293823508.htm (data obrashcheniya 01.08.2025).
6. Obzor gazoanalizatorov dlya metana / G.T. Zemskiy, A.V. Il'ichev, N.V. Kondratyuk, D.V. Dolgikh // Aktual'nye voprosy pozharnoy bezopasnosti. – 2021. – № 4(10). – S. 14–27. – DOI: 10.37657/vniipo.avpb.2021.35.86.003
7. Gas Sensor Drift Mitigation using Classifier Ensembles / A. Vergara, R. Huerta, T. Ayhan [et al.] // SensorKDD '11: Proceedings of the Fifth International Workshop on Knowledge Discovery from Sensor Data, San Diego, CA, USA, August 21, 2011. – P. 16–24. – DOI: 10.1145/2003653.2003655
8. Concept Drift Mitigation in Low-Cost Air Quality Monitoring Networks / G. D’Elia, M. Ferro, P. Sommella [et al.] // Sensors. – 2024. – Vol. 24, Issue 9. – P. 2786. – DOI: 10.3390/s24092786
9. Investigation of Electronic Nose Sensor Drift Correction Methods and Their Application to Environmental Samples / S. Vitali, C. Bax, S. Sironi, L. Capelli // Chemical Engineering Transactions. – 2018. – Vol. 68. – P. 289–294. – DOI: 10.3303/CET1868049
10. Samarin I.V., Strogonov A.Yu., Kryuchkov A.V. Model' prognozirovaniya klassa gotovnosti datchikov termokhimicheskogo gazosignalizatora dlya predotvrashcheniya pozharov i vzryvov // Pozharovzryvobezopasnost'. – 2024. – T. 33, № 5. – S. 87–98. – DOI: 10.22227/0869-7493.2024.33.05.87-98
11. Lin Juntao, Zhan Xianghao. Sensor Drift Compensation in Electronic-Nose-Based Gas Recognition Using Knowledge Distillation. – 2025. – DOI: 10.48550/arXiv.2507.17071
12. Gas Sensor Array Drift at Different Concentrations. – 2013. – URL: https://archive.ics.uci.edu/dataset/270/gas%2Bsensor%2Barray%2Bdrift%2Bdataset%2Bat%2Bdifferent%2Bconcentrations?utm (data obrashcheniya 04.08.2025).
13. Svidetel'stvo 2024684137 o gosudarstvennoy registratsii programmy dlya EVM. Programma generatsii sinteticheskogo nabora dannykh dlya resheniya s pomoshch'yu svertochnoy neyronnoy seti zadachi klassifikatsii sostoyaniy vynosnykh datchikov gazosignalizatorov na ob"ektakh toplivno-energeticheskogo kompleksa po dannym gazosignalizatorov STM-10 / A.Yu. Strogonov, A.V. Kryuchkov; pravoobladatel' FGAOU VO "RGU nefti i gaza (NIU) imeni I.M. Gubkina". – № 2024683595; zayavl. 15.10.2024; opubl. 15.10.2024, Byul. № 10.
14. Salikhov M.R., Yur'eva R.A. Algoritm prognozirovaniya sostoyaniya oborudovaniya na osnove mashinnogo obucheniya // Izvestiya vysshikh uchebnykh zavedeniy. Priborostroenie. – 2022. – T. 65, № 9. – S. 648–655. – DOI: 10.17586/0021-3454-2022-65-9-648-655
15. Samarin I.V., Kryuchkov A.V., Strogonov A.Yu. Raschet reglamentirovannogo kolichestva vynosnykh datchikov gazosignalizatorov vokrug otkrytykh ustanovok neftepererabatyvayushchikh zavodov pryamougol'noy formy // Pozharovzryvobezopasnost'. – 2022. – T. 31, № 2. – S. 33–40. – DOI: 10.22227/0869-7493.2022.31.02.33-40
16. Samarin I.V., Kryuchkov A.V., Strogonov A.Yu. Podkhod k modelirovaniyu tekhnicheskogo obsluzhivaniya statsionarnykh termokhimicheskikh gazosignalizatorov na ustanovkakh NPZ // Pozharovzryvobezopasnost'. – 2022. – T. 31, № 1. – S. 40–48. – DOI: 10.22227/0869-7493.2022.31.01.40-48
17. Samarin I.V., Kryuchkov A.V., Strogonov A.Yu. Model' otsenki gotovnosti termokhimicheskikh gazoanalizatorov // Pozharovzryvobezopasnost'. – 2020. – T. 29, № 6. – S. 61–74. – DOI: 10.22227/PVB.2020.29.06.61-74
18. Samarin I.V., Kruchkov A.V., Strogonov A.Yu. Creation of principles for the implementation of operating systems for high-precision fire equipment // Periodicals of Engineering and Natural Sciences. – 2021. – Vol. 9, No. 4. – P. 589–604. – DOI: 10.21533/pen.v9.i4.969