«Automation and Informatization of the fuel and energy complex»

Differentiable physics – the basis of digital twins in the oil and gas complex

UDC: 519.673
DOI: -

Authors:

¹ NIIgazekonomika, Moscow, Russia

Keywords: digital modeling, differentiable programming, hybrid model, optimization, technological process, system

Annotation:

The author of the article considers the experience and prospects for further implementation of a new paradigm – differentiable programming – into the practice of developing applied digital models of hydrocarbons production, transportation, storage, processing and distribution. It is argued that autodifferentiable models developed on the basis of differentiable programming methods (models of "differentiable physics") open new opportunities for improving the management of oil and gas production. Autodifferentiability as a property of digital computational models is an important component for solving various direct and inverse problems of detailed physical and technological modeling of oil and gas complex systems, including the problems of identifying the state of systems based on actual measurements and optimal planning of stationary and dynamic modes of their operation. It is noted that differentiable programming allows combining classical physical and mathematical models of filtration and fluid flow with machine learning models into integrated hybrid autodifferentiable models. It makes it possible to use the advantages of both knowledge-based and data-based models when modeling large systems. It is shown that the use of these approaches allows expanding the range of topical problems to be solved as well as greatly accelerating calculations due to the use of specialized hardware (graphics processors). The implementation of autodifferentiable hybrid models using modern artificial intelligence software packages expands the range of methods and tools for modeling and optimizing oil and gas systems. It is assumed that practical implementation of the proposed approaches based on differentiable programming will contribute to more intensive development of digital twins of oil and gas production. Examples of using the proposed approaches to solve applied industry problems are given. The challenges that require overcoming when expanding the use of the new paradigm are noted.

Bibliography:

1. Razrabotka i aprobatsiya metodicheskikh podkhodov i tsifrovykh tekhnologiy neyrosetevogo proksi-modelirovaniya ustanovivshegosya dvukhfaznogo techeniya mnogokomponentnoy smesi v sistemakh sbora i promyslovoy podgotovki gaza (na primere Chayandinskogo NGKM) / A.V. Belinskiy, V.A. Marishkin, V.V. Samsonova, P.V. Pyatibratov // Avtomatizatsiya i informatizatsiya TEK. – 2024. – № 4(609). – S. 44–59.
2. 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
3. Kislenko N.A., Belinskiy A.V., Kazak A.S. Metody, algoritmy i instrumenty modelirovaniya i optimizatsii rezhimov raboty Edinoy sistemy gazosnabzheniya Rossii na osnove tekhnologiy iskusstvennogo intellekta. Chast' 1 // Gazovaya prom-st'. – 2021. – № 9(821). – S. 88–96.
4. Kislenko N.A., Belinskiy A.V., Kazak A.S. Metody, algoritmy i instrumenty modelirovaniya i optimizatsii rezhimov raboty Edinoy sistemy gazosnabzheniya Rossii na osnove tekhnologiy iskusstvennogo intellekta. Chast' 2 // Gazovaya prom-st'. – 2021. – № 10(822). – S. 98–104.
5. Sukharev M.G., Popov R.V. Sostoyanie i perspektivy sovershenstvovaniya matematicheskogo i komp'yuternogo obespecheniya v sfere operativnogo upravleniya i srednesrochnogo planirovaniya rezhimov krupnomasshtabnykh gazotransportnykh sistem // Nauch.-tekhn. sb. Vesti gazovoy nauki. – 2018. – № 2(34). – S. 4–13.
6. Motaei E., Ganat T. Smart proxy models art and future directions in the oil and gas industry: A review // Geoenergy Science and Engineering. – 2023. – Vol. 227. – P. 211918. – DOI: 10.1016/j.geoen.2023.211918
7. Blondel M., Roulet V. The Elements of Differentiable Programming. – 2024. – 451 p. – DOI: 10.48550/arXiv.2403.14606
8. Automatic differentiation in machine learning: A survey / A.G. Baydin, B.A. Pearlmutter, A.A. Radul, J.M. Siskind // J. of Machine Learning Research. – 2018. – Vol. 18, Issue 153. – P. 1–43. – DOI: 10.48550/arXiv.1502.05767
9. Griewank A. On automatic differentiation // Mathematical Programming: Recent Developments and Applications / editors: M. Iri, K. Tanabe. – Kluwer Academic Publishers, 1989. – P. 83–108.
10. Ramsundar V., Krishnamurthy D., Viswanathan V. Differentiable physics: A position piece. – 2021. – 12 p. – DOI: 10.48550/arXiv.2109.07573
11. Sardanashvili S.A. Raschetnye metody i algoritmy (truboprovodnyy transport gaza). – M.: Izd-vo "Neft' i gaz" RGU nefti i gaza im. I.M. Gubkina, 2005. – 577 s.
12. Griewank A., Walther A. Evaluating Derivatives: Principles and Techniques of Algorithmic Differentiation. – Philadelphia: Society for Industrial and Applied Mathematics, 2008. – XXII, 438 p. – DOI: 10.1137/1.9780898717761
13. O novom metode tsifrovogo modelirovaniya nestatsionarnykh rezhimov techeniya gaza v magistral'nykh gazoprovodakh s primeneniem neyronnykh operatorov / A.V. Belinskiy, D.V. Gorlov, I.A. Pyatyshev, A.E. Titov // Gazovaya prom-st'. – 2024. – № 5(865). – S. 54–66.
14. Sukharev M.G., Samoylov R.V. Analiz i upravlenie statsionarnymi i nestatsionarnymi rezhimami transporta gaza. – M.: Izdat. tsentr RGU nefti i gaza (NIU) im. I.M. Gubkina, 2017. – 399 s.
15. Sukharev M.G., Popov R.V. Novaya metodika modelirovaniya nestatsionarnykh techeniy gaza v sistemakh gazosnabzheniya // Izv. RAN. Energetika. – 2015. – № 2. – S. 150–159.