«Automation and Informatization of the fuel and energy complex»

Approximation of gas compressibility factor based on genetic algorithms

UDC: 004.023+004.89+622.691.4
DOI: 10.33285/2782-604X-2023-11(604)-59-68

Authors:

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

Keywords: natural gas, equation of state, compressibility coefficient, genetic programming, surrogate modeling

Annotation:

For hydraulic calculations of gas flow parameters in a gas pipeline it is necessary to calculate a compressibility factor z(p,T), which allows taking into account the deviation of real gas behavior from the ideal one. A large number of approximation dependences in explicit and implicit form, based on the accumulated significant amount of experimental data, have been developed. The calculation of the compressibility factor, taking into account the gas component composition can be performed on the basis of the equation of state (in recent years, the majority of researchers rely on GERG-2008, AGA8, AGA10), where the problem comes down to the iterative method of solving a nonlinear equation, the same procedure is proposed in [1]. To model non-stationary modes of gas transmission in trunk pipelines, to determine the optimal gas transportation mode and to identify parameters of a gas transportation system, the time, spent on calculation is of great importance. The aim of the work was to use the method of genetic programming (another name of the method is symbolic regression) for building approximation dependences for the compressibility factor in the explicit form as well as to analyze the quality of the obtained models.

Bibliography:

1. GOST 30319.3-2015. Gaz prirodnyy. Metody rascheta fizicheskikh svoystv. Vychislenie fizicheskikh svoystv na osnove dannykh o komponentnom sostave. – Vved. 2017–01–01. – M.: Standartinform, 2016. – III, 28 s.
2. Aleksanochkin A.A., Sardanashvili S.A. Raschetnye metody opredeleniya svoystv prirodnogo gaza po dannym o plotnosti pri standartnykh usloviyakh, soderzhanii azota i dioksida ugleroda // Delovoy zhurn. Neftegaz.RU. – 2020. – № 1(97). – S. 78–85.
3. Aleksanochkin A.A., Sardanashvili S.A. Raschetnye metody opredeleniya fizicheskikh i termodinamicheskikh svoystv prirodnogo gaza. Metod razlozheniya nepolnogo komponentnogo sostava prirodnogo gaza na ekvivalentnyy komponentnyy sostav // Tr. RGU nefti i gaza im. I.M. Gubkina. – 2018. – № 3(292). – S. 184–194.
4. Azizi N., Behbahani R.M., Isazadeh M.A. An efficient correlation for calculating compressibility factor of natural gases // J. of Natural Gas Chemistry. – 2010. – Vol. 19, Issue 6. – P. 642–645. – DOI: 10.1016/S1003-9953(09)60081-5
5. Kareem L.A., Iwalewa T.M., Al-Marhoun M. New explicit correlation for the compressibility factor of natural gas: linearized z-factor isotherms // J. of Petroleum Exploration and Production Technology. – 2016. – Vol. 6. – P. 481–492. – DOI: 10.1007/s13202-015-0209-3
6. Hall K.R., Yarborough L. A new equation of state for Z-factor calculations // Oil Gas J. – 1973. – Vol. 71, Issue 25. – P. 82–92.
7. A novel efficient model for gas compressibility factor based on GMDH network / Luan Lin, Shiyang Li, Sihao Sun [et al.] // Flow Measurement and Instrumentation. – 2020. – Vol. 71. – P. 101677. – DOI: 10.1016/j.flowmeasinst.2019.101677
8. Azizi N., Rezakazemi M., Zarei M.M. An intelligent approach to predict gas compressibility factor using neural network model // Neural Computing and Applications. – 2019. – Vol. 31. – P. 55–64. – DOI: 10.1007/s00521-017-2979-7
9. Koza J.R. Genetic programming: on the programming of computers by means of natural selection. – Cambridge, Mass.: The MIT Press, 1992. – XIV, 819 p.
10. Searson D.P., Leahy D.E., Willis M.J. Predicting the toxicity of chemical compounds using GPTIPS: A free genetic programming toolbox for MATLAB // Lecture Notes in Electrical Engineering. – 2011. – Vol. 70. Intelligent Control and Computer Engineering. – P. 83–93. – DOI: 10.1007/978-94-007-0286-8_8
11. Gandomi A.H., Alavi A.H. A new multi-gene genetic programming approach to non-linear system modeling. Part II: geotechnical and earthquake engineering problems // Neural Computing and Applications. – 2012. – Vol. 21. – P. 189–201. – DOI: 10.1007/s00521-011-0735-y
12. Praks P., Brkic D. Symbolic Regression-Based Genetic Approximations of the Colebrook Equation for Flow Friction // Water. – 2018. – Vol. 10, Issue 9. – P. 1175. – DOI: 10.3390/w10091175
13. Kochueva O.N. Feature Selection for a Fuzzy Classification Model Based on a Genetic Algorithm // Raspredelennye komp'yuternye i telekommunikatsionnye seti: upravlenie, vychislenie, svyaz' (DCCN-2022) = Distributed computer and communication networks: control, computation, communications (DCCN-2022): materialy XXV Mezhdunar. nauch. konf., M., 26–30 sent. 2022 g. – M.: Rossiyskiy universitet druzhby narodov (RUDN), 2022. – S. 168–175.
14. Kochueva O.N. Razrabotka modeley prognozirovaniya vybrosov oksidov ugleroda i azota gazovykh turbin na osnove geneticheskikh algoritmov // Delovoy zhurn. Neftegaz.RU. – 2022. – № 5-6(125-126). – S. 14–20.
15. Jinbo Bi, Bennett K.P. Regression Error Characteristic Curves // Proc. of the 20th Int. Conf. on Machine Learning (ICML-2003), Aug. 21–24, 2003, Washington, DC, USA. – 2003. – P. 43–50.