A.R. Mukhutdinov1, M. G. Efimov2, A.N. Serdyukova3

1-3 Kazan National Research Technological University (Kazan, Republic of Tatarstan)

It is known that for warming up and creating a high temperature in the zone of a productive formation, as well as initiating in-situ combustion (IG), liquid-filled compositions (LFS) are used. There is a lot of information in the literature regarding the LFS for the initiation of IG, optimal formulations are given. However, their inherent disadvantages (the extraction process is long, complicated and expensive) limit their use. In this regard, it is important to find the optimal calorific value of the LFS with stable ignition and complete combustion. A promising way to find the optimal LFS is to use the universal computing capabilities of modern software tools for the development of artificial neural networks (ANNs), which have the widest possibilities for modeling such systems. Therefore, the determination of the optimal LFS through the use of ANN to achieve the maximum calorific value with its stable combustion is of scientific and practical interest.

A large array of experimental data in the scientific literature concerning thermal methods of influencing the formation during in-situ combustion is fragmented and little studied. Therefore, the main goal of the work is to determine the optimal LFS, which has the maximum calorie content, through the use of the developed PM based on the ANN and the formed knowledge base from the processed and systematized information.

In this paper, a knowledge base and an ANN with general regression using a non-linear Gaussian-type activation function based on it, as well as an applied PM, were developed in order to find the optimal caloric value of the LFS to ensure the intensification of oil inflow by heating a bituminous reservoir. Based on the developed and pre-trained ANN with a relative error of no more than 2,3%, the forecast of the output parameter, in this case, the calorie content of the optimal LFS, was clearly demonstrated.

The developed applied PM based on the ANN is able to determine the optimal calorie content of the oil, which will ensure the intensification of the inflow of high-viscosity hard-to-recover bituminous oil.

Mukhutdinov A.R., Efimov M.G., Serdyukova A.N. Neural network optimization model for the calories of the liquid-filled composition for internal production bituminous oil. Nonlinear World. 2023. V. 21. № 1. P. 21-26. DOI: https://doi.org/10.18127/j20700970-202301-03 (In Russian)

1. Muslimov R.H., Romanov G.V., Kajukova G.P., Jusupova T.N., Iskrickaja N.I., Petrov S.M. Strategija razvitija neftebitumnogo kompleksa Tatarstana v napravlenii vosproizvodstva resursnoj bazy uglevodorodov. Neft'. Gaz. Novacii. 2012. № 2. S. 21-29 (In Russian).
2. Kudinov V.I. Novye tehnologii povyshenija nefteotdachi na mestorozhdenijah s vysokovjazkimi neftjami. Neftjanoe hozjajstvo. 2002. № 5. S. 92-95 (In Russian).
3. Gupta P., Doriah A., Rjej S. Rezul'taty vnutriplastovogo gorenija. Neftegazovye tehnologii. 2008. № 3. S. 12-15 (In Russian).
4. Pyresev V.G., Sharnina A.P., Vavilov Ju.G., Kudinov V.I. Razrabotka izdelij dlja progreva bituminoznogo plasta: Diplomnaja rabota. Kazan': KHTI im. S.M. Kirova. 1972 (In Russian).
5. Sadykov I.F., Frolov G.P. Otchet po NIR «Razrabotka i ispytanie zhidkonapolnennyh sostavov dlja progreva bitumov i iniciirovanija VG». Kazan': KHTI im. S.M. Kirova. 1989 (In Russian).
6. Otchet po NIR «Ispytanie v promyslovyh uslovijah, usovershenstvovannyh zhidkonapolnennyh sostavov dlja iniciirovanija VG i progreva bitumnogo plasta». Kazan': KHTI im. S.M. Kirova. 1990 (In Russian).
7. Bochkov V.M., Sadykov I.F., Sheshukov G.I., Frolov G.P., Sabitov I.M., Gajnutdinov I.A. Otchet po NIR «Poiskovye issledovanija po sozdaniju izdelij dlja progreva bituminoznogo plasta». Kazan': KHTI im. S.M. Kirova. S. 6-24 (In Russian).
8. Liquid Explosives. College of Materials Science and Engineering, Beijing Institute of Technology. Springer: Jiping Liu. 2015.
9. Zhang C., Chen W., Wang Z., et al. Research progress of azide polymer adhesives plasticized by nitric acid ester. China Adhes. 2012.
10. Formulation and characterization of a new nitroglycerin-free double base propellant. Propellants Explos Pyrotech. Reese DA. Groven LJ. Son SF. 2014.
11. De Tata D.A., Fillingham R.M., Dunsmore R.P. Encyclopedia of Forensic Sciences. 3rd Edition. Commercial and Military Explosives. 2023. V. 1. P. 605-621.
12. Muhutdinov A.R., Marchenko G.N., Vahidova Z.R. Nejrosetevoe modelirovanie i optimizacija slozhnyh processov i naukoemkogo teplojenergeticheskogo oborudovanija. Kazan': Kazanskij gos. jenergeticheskij un-t. 2011. 296 s. (In Russian)
13. Muhutdinov A.R., Efimov M.G. Nejrosetevoj podhod dlja optimizacii sostava tverdogo topliva po skorosti gorenija. Avtomatizacija, telemehanizacija i svjaz' v neftjanoj promyshlennosti. 2019. № 4. S. 25-29 (In Russian).
14. Muhutdinov A.R., Sadykov M.I., Efimov M.G. Optimizacija receptury tverdogo topliva s ispol'zovaniem komp'juternyh tehno-logij. Strategicheskaja stabil'nost'. 2018. № 4. S. 64-66 (In Russian).
15. Muhutdinov A.R., Efimov M.G. Universal'nye vychislitel'nye jekspress-metody dlja sozdanija iskusstvennoj nej-ronnoj seti slozhnogo ob’ekta i innovacionnogo programmnogo modulja na ee osnove. Monografija. Kazan': Izd-vo KNITU. 2022. 164 s. (In Russian).