«Automation and Informatization of the fuel and energy complex»

Forecasting of oil quality indicators using the SARIMAX model, a polynomial and the trigonometric function

UDC: 681.5.08:004.942
DOI: -

Authors:

¹ Almetyevsk State Technological University "Higher School of Petroleum", Almetyevsk, Russia
² Nefteavtomatika, Almetyevsk, Russia

Keywords: SARIMAX, mathematical modeling, time series forecasting, oil industry, seasonal fluctuations, trend, data analysis, oil quality control, industrial data

Annotation:

The authors of the article reviewed the methods applied for modeling and forecasting the parameters of the processes, going on in the fuel and energy complex. The object of the research is the installation for measuring the quantity and quality indicators of oil. The authors present a time series model based on ARIMAX and SARIMAX for predicting oil moisture content. The SARIMAX model takes into account seasonality and external factors. The average relative error of approximation was 8,24 %, and the average relative error of post-prediction made 9,38 %. The model has been improved by using a trend in the form of a polynomial and a trigonometric function. After subtracting the trend, the seasonality and the residual are modeled by SARIMAX. A distinctive feature of the proposed model is the fact that it predicts not the function itself with constant coefficients, but the coefficients of the function. The average relative error of approximation was 1,12 % and the average relative error of post-prediction made 7,12 %. Loginom and Python software tools have been selected as the tools for implementing the models. The study was conducted using real data. It has been shown that accounting for seasonal fluctuations and external factors (temperature, density) as well as representing the trend as a sum of a polynomial and trigonometric functions allows reducing the average relative error of the post-prediction. The use of mathematical modeling in combination with data analysis tools demonstrates the potential for integrating such solutions into automated systems for monitoring the quality of the fuel and energy complex.

Bibliography:

1. Rudin M.G., Somov V.E., Fomin A.S. Karmannyy spravochnik neftepererabotchika. – M.: TsNITTEneftekhim, 2004. – 336 s.
2. Afanas'ev V.N., Yuzbashev M.M. Analiz vremennykh ryadov i prognozirovanie. – M.: Finansy i statistika, 2001. – 226 s.
3. Kizina I.D., Verevkin A.P. Metody verifikatsii i dostoverizatsii dannykh dlya resheniya zadach v avtomatizirovannykh sistemakh upravleniya tekhnologicheskimi protsessami neftegazovoy otrasli // Avtomatizatsiya i informatizatsiya TEK. – 2025. – № 1(618). – S. 42–47.
4. Golyandina N.E. Metod "Gusenitsa" – SSA: analiz vremennykh ryadov: uchebnoe posobie. – SPb.: SPbGU, 2004. – 76 s.
5. Prognozirovanie elektropotrebleniya na osnove metoda glavnykh komponent (PCA) / I.U. Rakhmonov, N.N. Niezov, V.Ya. Ushakov [i dr.] // Izvestiya Tomskogo politekhnicheskogo universiteta. Inzhiniring georesursov. – 2024. – T. 335, № 12. – S. 198–209. – DOI: 10.18799/24131830/2024/12/4731
6. Bandara K., Hyndman R.J., Bergmeir C. MSTL: A Seasonal-Trend Decomposition Algorithm for Time Series with Multiple Seasonal Patterns // International Journal of Operational Research. – 2025. – Vol. 52, No. 1. – P. 79–98. – DOI: 10.1504/IJOR.2025.143957
7. Tugashova L.G., Zatonskiy A.V. Modelirovanie i prognozirovanie proizvodstva nefteproduktov s uchetom sezonnosti na osnove avtoregressionnykh modeley // Izvestiya Tomskogo politekhnicheskogo universiteta. Inzhiniring georesursov. – 2020. – T. 331, № 5. – S. 109–119. – DOI: 10.18799/24131830/2020/5/2641
8. Tugashova L.G. Prognozirovanie pokazateley kachestva nefteproduktov na ustanovkakh pervichnoy pererabotki nefti // Uchenye zapiski Al'met'evskogo gosudarstvennogo neftyanogo instituta. – 2015. – T. 14. – S. 99–103.
9. Model' dlya prognozirovaniya poverkhnostnoy kontsentratsii metana v arkticheskom regione, osnovannaya na iskusstvennoy neyronnoy seti s dlinnoy tsep'yu elementov kratkosrochnoy pamyati i veyvlet-preobrazovaniem iskhodnykh dannykh / A.G. Buevich, A.P. Sergeev, A.V. Shichkin [i dr.] // Arktika: ekologiya i ekonomika. – 2023. – T. 13, № 3(41). – S. 428–436. – DOI: 10.25283/2223-4594-2023-3-428-436