«Automation and Informatization of the fuel and energy complex»

Integration of neural network models and expert knowledge for early fault detection in operation of gas pumping units

UDC: 681.5:622.279
DOI: 10.33285/2782-604X-2023-10(603)-5-16

Authors:

¹ Gazprom transgaz Chaikovsky, Chaikovsky, Russia
² NIIgazeconomika, Moscow, Russia

Keywords: natural gas transportation, gas pumping units, predictive analytics, neural networks

Annotation:

Ensuring the reliability and stability of gas supplies to consumers requires using a wide range of methods for managing the technical state of the operated power equipment of the gas transmission system (GTS). Modern digital technologies open up new opportunities for early detection of defects and malfunctions when operating equipment in order to make timely management decisions. The article proposes methodological approaches to building computer models designed to detect the emergence of unfavorable trends in the operation of gas compressor units (GCU), the implementation of which can lead to malfunctions and failures. A feature of the proposed models is the use of a neural network apparatus in them, based on the expert knowledge in the field of operation of the power equipment of the gas transmission system (GTS). The results of computational experiments based on the processing of a representative sample of data containing time series of technological parameters of the operation of real gas compressor units are presented. The advantages and disadvantages of the developed models are mentioned, the ways of their further development are outlined.

Bibliography:

1. GOST 13377-75. Nadezhnost' v tekhnike. Terminy. – M.: Izd-vo standartov, 1975. – 21 s.
2. Sverdlov A.B. Povyshenie nadezhnosti gazoperekachivayushchikh agregatov putem primeneniya tekhnologii epilamirovaniya // Nauch.-tekhn. vedomosti S.-Peterb. gos. politekhn. un-ta. – 2014. – № 3(202). – S. 62–69.
3. Paklin N.B., Oreshkov V.I. Biznes-analitika: ot dannykh k znaniyam. – SPb.: Piter, 2016. – 704 s.
4. Oleynikov A.V., Kazak A.S. Sovremennoe sostoyanie i perspektivy tsifrovoy transformatsii v neftegazovykh kompaniyakh // Avtomatizatsiya i informatizatsiya TEK. – 2023. – № 3(596). – S. 12–27. – DOI: 10.33285/2782-604X-2023-3(596)-12-27
5. Zaritskiy S.P. Diagnostika gazoperekachivayushchikh agregatov gazoturbinnymi privodami. – M.: Nedra, 1987. – 198 s.
6. Paklin N.B., Oreshkov V.I. Biznes-analitika: ot dannykh k znaniyam. – SPb.: Piter, 2009. – 624 s.
7. Gafarov F.M., Galimyanov A.F. Iskusstvennye neyronnye seti i prilozheniya: ucheb. posobie. – Kazan': Izd-vo Kazan. un-ta, 2018. – 121 s.
8. Girin R.V., Orlov S.P. Dvukhstadiynaya normalizatsiya vykhodnykh signalov iskusstvennykh neyronnykh setey // Vestn. SamGTU. Ser.: Tekhnicheskie nauki. – 2017. – № 4(56). – S. 7–16.
9. Ob obosnovannosti primeneniya, sovremennom sostoyanii i nekotorykh perspektivakh razvitiya neyrosetevykh modeley Edinoy sistemy gazosnabzheniya Rossii / N.A. Kislenko, A.V. Belinskiy, A.S. Kazak, O.I. Belinskaya // Avtomatizatsiya i informatizatsiya TEK. – 2022. – № 5(586). – S. 6–17. – DOI: 10.33285/2782-604X-2022-5(586)-6-17
10. Belinskiy A.V. Tsifrovye tekhnologii modelirovaniya gazotransportnykh sistem // GAZPROM: Korporativnyy zhurn. PAO "Gazprom". – 2022. – № 12. – S. 40–47.
11. Opyt i perspektivy primeneniya matematicheskogo apparata, vychislitel'nykh algoritmov i komp'yuternykh tekhnologiy mashinnogo obucheniya v zadachakh issledovaniya, funktsionirovaniya i razvitiya sistem gazosnabzheniya / N.A. Kislenko, A.V. Belinskiy, A.S. Kazak, E.V. Fomina // Matematicheskie modeli i metody analiza i optimal'nogo sinteza razvivayushchikhsya truboprovodnykh i gidravlicheskikh sistem: tr. XVIII Vseros. nauchn. seminara, Irkutsk, 12–18 sent. 2022 g. – Irkutsk: ISEM SO RAN, 2022. – S. 39–55.