Neural network system for automated control of thermochemical dehydration of oil emulsions

UDK: 665.622.43.066.6: 004.942
DOI: 10.24887/0028-2448-2019-6-102-105
Key words: intelligent field, neural network, modeling, emulsion, oil dehydration
Authors: I.V. Artyushkin (Giprovostokneft JSC, RF, Samara), G.N. Rogachev (Samara State Technical University, RF, Samara), V.N. Yakimov (Samara State Technical University, RF, Samara), E.E. Yaroslavkina (Samara State Technical University, RF, Samara)

This paper addresses the issue of establishing intelligent technical complex employing real time adaptive control of oil dehydration process. For this purpose functional model of oil dehydration process was developed to present emulsified crude oil as a non-linear multi-dimensional and multi-loop controlled object. Such model became the basis for the development of adaptive automated control system. The system incorporates three adaptive loops. The first loop adjusts control activity in accordance with the response of reference model. The second loop adjusts object control in accordance with the disturbance. The third adaptive loop aggregates information on control command values and object response, for regular update of the reference model in the first loop. Artificial neural network was built for implementing smart control solutions. Neural network was trained using experience-based data from Oil Processing Recommendations by Reservoir Engineering Division of Giprovostokneft. Designed adaptive system enables monitoring of residual water content in oil as well as input parameters of dehydration process. It also calculates transport delay for emulsion breakdown, and enables process control in accordance with the design values of dehydration process model. Control architecture can be integrated into operating facilities or used in the new oil processing facilities. Individual reference model shall be prepared for each object considering properties and content of crude oil emulsion. Adaptive control can be part of the general control system of oil processing facility. The proposed system is one of the phases of unmanned technologies implementation allowing the use of self-contained unattended facilities.


1. Pozdnyshev G.N., Stabilizatsiya i razrushenie neftyanykh emul'siy (Stabilization and destruction of oil emulsions), Moscow: Nedra Publ., 1982, 221 p.

2. Tronov V.P., Promyslovaya podgotovka nefti (Field oil treatment), Kazan': FEN Publ., 2000. – 416 p.

3. Putokhin V.S., Matematicheskoe modelirovanie tekhnologicheskogo protsessa obezvozhivaniya nefti na promyslakh (Mathematical modeling of the technological process of oil dehydration in the fields), Collected papers “Neft' i gaz” (Oil and gas), Moscow: Publ. of . Moscow Institute of Petrochemical and Gas Industry named after I. M. Gubkin, 1977, pp. 37–42.

4. Verevkin A.P., El'tsov I.D., Zozulya Yu.I., Kiryushin O.V., Operational management of technological processes for oil treatment according to technical and economic indicators (In Russ.), Avtomatizatsiya, telemekhanizatsiya i svyaz' v neftyanoy promyshlennosti, 2006, no. 3, pp. 48–53.

5. Andreev E.B., Klyuchnikov A.I., Krotov A.V. et al., Avtomatizatsiya tekhnologicheskikh protsessov dobychi i podgotovki nefti i gaza (Automation of technological processes of oil and gas production and treatment), Moscow: Nedra-Biznestsentr Publ., 2008, 399 p.

6. Artyushkin I.V., Building possibility investigation of complex expert automated control system for oil treatment technological process (In Russ.), Neftyanoe khozyaystvo = Oil Industry, 2016, no. 6, pp. 29–31.

7. Artyushkin I.V., Maksimov A.E., Automated process control system design for thermochemical dehydration based on neural network (In Russ.), Vestnik Samarskogo gosudarstvennogo tekhnicheskogo universiteta. Ser. Tekhnicheskie nauki, 2017, no. 1 (53), pp. 7–15.

8. Bortnikov A.E., Kordik K.E., Savinykh A.V., Nitsin A.S., Some results of laboratory experiments on destruction of oil-water emulsions exposed to uniform electric field (In Russ.), Geologiya, geofizika i razrabotka neftyanykh i gazovykh mestorozhdeniy, 2013, no. 9, pp. 48–56.

9. Grzymala-Busse J.W., Mroczek T., Definability in mining incomplete data, Procedia Computer Science, 2016, V. 96, pp. 179–186.

10. Aksenov S.V., Novosel'tsev V.B., Organizatsiya i ispol'zovanie neyronnykh setey: Metody i tekhnologii (The organization and the use of neural networks: methods and technologies), Tomsk: Publ. of scientific and technical literature, 2006, 128 p.

11. Callan R., The essence of neural networks, Prentice Hall, 1998, 248 p.

12. Neural Network Software, About NeuroSolutions, URL:

13. Haykin S., Neural networks: A comprehensive foundation, Prentice-Hall, 1999, 874 p.

14. Rutkovskaya D., Pilin'skiy M., Rutkovskiy L., Neyronnye seti, geneticheskie algoritmy i nechetkie sistemy (Neural networks, genetic algorithms and fuzzy systems), Moscow: Goryachaya liniya –Telekom Publ., 2006, 452 p.

15. Stashkova O.V., Shestopal O.V., Use artificial neural networks for restoration of initial data array (In Russ.), Izvestiya vuzov. Severo-Kavkazskiy region. Ser. Tekhnicheskie nauki, 2017, no. 1, pp. 37–42.

16. Raimondi A., Favela-Contreras A., Beltrán-Carbajal F. et al., A design of an adaptive predictive control strategy for crude oil atmospheric distillation process, Control Engineering Practice, 2015, V. 34, pp. 39–48.

To buy the complete text of article (a format - PDF) or to read the material which is in open access only the authorized visitors of the website can. .

Mobile applications

Read our magazine on mobile devices

Загрузить в Google play

Press Releases