R.A. Batashov1

1 Plekhanov Russian University of Economics (Moscow, Russia)
1 rusbatashov@gmail.com

The paper considers the problem of forecasting petroleum product imbalance in tank accounting systems using a neural network module integrated into the information-measuring system. The relevance of the study is determined by the fact that, during receipt, storage, and delivery of petroleum products, the discrepancy between actual and accounting mass is formed under the influence of measurement errors in level, temperature, density, volume, and mass, as well as cumulative errors of the accounting process. A multilayer perceptron is proposed as an intelligent module for forecasting absolute imbalance based on technological and accounting features.

For numerical validation, an integrated data corpus was used, formed on the basis of operational tank accounting tables, automated measurements, interpolated and semi-synthetic observations. The target variable is the absolute imbalance expressed in kilograms. Two data corpora were considered, an automated measurement corpus and a combined corpus; for each corpus, reduced and extended feature sets were evaluated. The MLP model uses petroleum product level, temperature, density, mass, volume, and derived features characterizing changes in the measuring system state as input variables. The obtained results show that the neural network module can be used as a predictive element of an information-measuring system for subsequent selection of corrective action and reduction of imbalance within regulatory tolerance limits.

Batashov R.A. MLP model for forecasting petroleum product imbalance as a basis for adaptive correction in accounting systems // Nonlinear World. 2026. V. 24. № 3. P. 71–83. DOI: https:// doi.org/10.18127/ j20700970-202603-08

1. GOST R 8.595–2004. Gosudarstvennaya sistema obespecheniya edinstva izmerenij. Massa nefti i nefteproduktov. Obshchie trebovaniya k metodikam vypolneniya izmerenij. M.: Standartinform, 2004 (In Russian).
2. Godnev A.G., Zorya E.I. Teoriya i praktika izmerenij i ucheta kolichestva pri oborote nefteproduktov: monografiya. M.: MAKS Press. 2020. 272 s. DOI: 10.29003/m1393.978-5-317-06428-0 (In Russian).
3. Sistemnyj podhod k minimizacii debalansa nefteproduktov v rezervuarnyh parkah. Informacionno-izmeritel'nye i upravlyayushchie sistemy. 2023. T. 21. № 5. S. 37–51. DOI: 10.18127/j20700814-202305-02 (In Russian).
4. Batashov R.A. Issledovanie modeli iskusstvennyh nejronnyh setej pri obrabotke i analize dannyh s izmeritel'noj tekhniki. Informacionno-izmeritel'nye i upravlyayushchie sistemy. 2025. T. 23. № 1. S. 33–40. DOI: 10.18127/j20700814-202501-04 (In Russian).
5. Batashov R.A. Issledovanie modelej mashinnogo obucheniya v izmereniyah urovnya nefteproduktov emkostnymi urovnemerami. Informacionno-izmeritel'nye i upravlyayushchie sistemy. 2025. T. 23. № 3. S. 27–36 (In Russian).
6. Batashov R.A. Bajesovskij podhod k algoritmam korrekcii debalansa v informacionno-izmeritel'nyh sistemah ucheta nefteproduktov. Informacionno-izmeritel'nye i upravlyayushchie sistemy. 2026. T. 1. № 3. S. 10–19 (In Russian).
7. Bishop C.M. Pattern Recognition and Machine Learning. New York: Springer, 2006. 738 p.
8. James G., Witten D., Hastie T., Tibshirani R. An Introduction to Statistical Learning: with Applications in R. 2nd ed. New York: Springer. 2021. 607 p.
9. Godnev A.G., Zorya E.I., Nesgovorov D.A., Davydov N.V. Kommercheskij uchet tovarnyh potokov nefteproduktov avtomatizirovannymi sistemami. M.: MAKS Press. 2008. 426 s. (In Russian).
10. Zorya E.I., Klejner G.B., Skripnikov A.V., Cagareli D.V. Neft'–Toplivo–Ekonomika. Situaciya, problemy, perspektivy. M.: IC «Matematika». 1996. 231 s. (In Russian)
11. Zorya E.I., Korolenok A.M., Loshchenkova O.V., Kitashov Yu.N. Osnovy resursosberezheniya pri oborote uglevodorodov. M.: MAKS Press. 2018. 640 s. (In Russian).
12. Zorya E.I., Zenin V.I., Nikitin O.V., Prohorov A.D. Resursosberegayushchij servis nefteproduktoobespecheniya. M.: FGUP Izd-vo «Neft' i gaz» RGU nefti i gaza im. I.M. Gubkina. 2004. 448 s. (In Russian).
13. Hanov N.I., Fathutdinov A.Sh., Slepyan M.A., Zolotuhin E.A., Fathutdinov T.A., Kolovertnov G.Yu. Izmereniya kolichestva i kachestva nefti i nefteproduktov pri sbore, transportirovke, pererabotke i kommercheskom uchete. SPb.: Izd-vo SPbUEF. 2000. 270 s. (In Russian).
14. Zorya E.I., Godnev A.G., Nikulin A.E. Priem nefteproduktov ot postavshchikov po kolichestvu i kachestvu. Prakticheskoe posobie. M.: ZAO «Biznes Proekt». 2006. 340 s. (In Russian).
15. Godnev A.G. Shirokodiapazonnyj diskretno-nepreryvnyj datchik urovnya. Sistemnyj analiz, upravlenie i obrabotka informacii v kosmicheskoj otrasli. 2015. № 3. S. 189–194 (In Russian).
16. He X., Zhao K., Chu X. AutoML: A Survey of the State-of-the-Art. Knowledge-Based Systems. 2021. V. 212. Art. 106622. DOI: 10.1016/j.knosys.2020.106622.
17. Hutter F., Kotthoff L., Vanschoren J. Automated Machine Learning: Methods, Systems, Challenges. Cham: Springer. 2019. 219 p. DOI: 10.1007/978-3-030-05318-5.
18. Feurer M., Hutter F. Hyperparameter Optimization. Automated Machine Learning: Methods, Systems, Challenges. Cham: Springer. 2019. P. 3–33.
19. Gelman A., Carlin J.B., Stern H.S., Dunson D.B., Vehtari A., Rubin D.B. Bayesian Data Analysis. 3rd ed. Boca Raton: Chapman and Hall/CRC. 2013. 675 p.
20. Kalman R.E. A New Approach to Linear Filtering and Prediction Problems. Journal of Basic Engineering. 1960. V. 82. № 1. P. 35–45.