«Automation and Informatization of the fuel and energy complex»

Forecasting oil and liquid inlet at the oil collection station of an oil and gas producing enterprise using machine learning methods

UDC: 681.5:622.276
DOI: -

Authors:

¹ SamaraNIPIneft, Samara, Russia
² SamSTU, Samara, Russia

Keywords: forecasting system, machine learning methods, decision-making trees, neural network

Annotation:

The authors of the article consider machine learning methods and a recurrent neural network for developing an information system for forecasting oil and liquid production and inflow to the oil collection and treatment station of an oil and gas producing enterprise. Linear regression and XGBOOST decision tree ensembles are analyzed. In the course of the project, regression models such as linear regression, K-nearest neighbors (KNN), decision tree ensembles (XGBOOST), and artificial neural network (ANN)-based models – multilayer linear perceptron (MLP) and LSTM were studied and developed. The specified models and their responses to changes in global parameters were investigated.

Bibliography:

1. Silaev K.O., Silaeva A.N. Metody dlya analiza krivykh padeniya dobychi, primenyaemye v toplivno-energeticheskoy sfere // Ekonomika i sotsium. – 2016. – № 9(28). – S. 543–546.
2. Tung L.V. Primenenie metoda krivoy padeniya Arpsa dlya prognoza dobychi nefti skvazhiny N1 mestorozhdeniya "Chernyy Drakon", V'etnam // Problemy geologii i osvoeniya nedr: tr. XXII Mezhdunar. simp. im. akad. M.A. Usova studentov i molodykh uchenykh, posvyashch. 155-letiyu so dnya rozhdeniya akad. V.A. Obrucheva, 135-letiyu so dnya rozhdeniya akad. M.A. Usova, osnovateley Sibirskoy gorno-geologicheskoy shk., i 110-letiyu pervogo vypuska gornykh inzhenerov v Sibiri, Tomsk, 02–07 apr. 2018 g.: v 2 t. T. 2. – Tomsk: TPU, 2018. – S. 148–149.
3. Srednesrochnoe prognozirovanie dobychi nefti po mestorozhdeniyu na osnove modeley SARIMAX / A.F. Azbukhanov, I.A. Lakman, A.A. Agapitov, L.F. Sadikova // Neft. khoz-vo. – 2020. – № 2. – S. 84–88. – DOI: 10.24887/0028-2448-2020-2-84-88
4. Galkin V.I., Repina V.A. Razrabotka statisticheskoy modeli prognozirovaniya debita nefti na primere plasta Tl2a odnogo iz mestorozhdeniy Bashkirskogo svoda // Neft. khoz-vo. – 2015. – № 11. – S. 111–113.
5. Zinatullina A.M. Model' kratkosrochnogo prognozirovaniya dobychi pri razrabotke plastov nizhnemelovogo vozrasta na osnove instrumentov analiza vremennykh ryadov: magistr. dis. – Tomsk, 2020. – 104 s. – URL: https://earchive.tpu.ru/handle/11683/62453
6. Semenychev V.K., Kurkin E.I. ARMA-modelirovanie urovnya godovoy dobychi nefti iz plasta i otsenka geologicheskogo riska investitsiy v neftegazodobyvayushchey promyshlennosti // Vestn. Samar. munitsip. in-ta upravleniya. – 2010. – № 2(13). – S. 7–15.
7. Tignibidin A.V., Panin Yu.N., Rusanova A.D. Snizhenie ekspluatatsionnykh zatrat posredstvom prognozirovaniya vneplanovykh prostoev skvazhin s pomoshch'yu mashinnogo obucheniya // Tekhnika i tekhnologiya neftekhimicheskogo i neftegazovogo proizvodstva: materialy 10-y Mezhdun. nauch.-tekhn. konf., Omsk, 26–29 fevr. 2020 g. – Omsk: OGTU, 2020. – S. 253–254.
8. Pichugin O.N., Prokof'eva Yu.Z., Aleksandrov D.M. Derev'ya resheniy kak effektivnyy metod analiza i prognozirovaniya // Neftepromyslovoe delo. – 2013. – № 11. – S. 69–75.
9. Krasnov F.V., Glavnov N.G., Sitnikov A.N. Primenenie mashinnogo obucheniya po ansamblyu reshayushchikh pravil dlya vychisleniya prognoza dopolnitel'nogo koeffitsienta izvlecheniya nefti // Int. J. of Open Information Technologies. – 2017. – T. 5, № 10. – S. 18–23.
10. Shippen M.E., Scott S.L. A Neural Network Model for Prediction of Liquid Holdup in Two-Phase Horizontal Flow // SPE Production & Facilities. – 2004. – Vol. 19, Issue 02. – P. 67–76. – DOI: 10.2118/87682-pa
11. Andrianov N. A Machine Learning Approach for Virtual Flow Metering and Forecasting // IFAC PapersOnLine. – 2018. – Vol. 51, Issue 8. – P. 191–196. – DOI: 10.1016/j.ifacol.2018.06.376
12. Farahani P.F., Bayat A.E., Rafati R. A Novel Method to Predict the Well Inflow Performance Relationships Curves by Artificial Intelligence Techniques // Int. J. of Petroleum and Petrochemical Engineering. – 2018. – Vol. 4, Issue 3. – P. 13–20. – DOI: 10.20431/2454-7980.0403002
13. Looking Back 2006 – 2016 – Ten Years of Data Driven Well Rate Estimates for Real-Time Surveillance and Optimization / J. Briers, Keat-Choon Goh, A. Sniekers [et al.] // SPE Intelligent Energy Int. Conf. and Exhibition, Aberdeen, Scotland, UK, Sept. 6–8, 2016. – P. 284–295. – DOI: 10.2118/181032-MS
14. Sun Jianlei, Ma X., Kazi M. Comparison of Decline Curve Analysis DCA with Recursive Neural Networks RNN for Production Forecast of Multiple Wells // SPE Western Regional Meeting, Garden Grove, California, USA, April 22–26, 2018. – DOI: 10.2118/190104-MS
15. A neural network approach to predict existing and in-fill oil well performance / Linyu Yang, Zhong He, J. Yen, Ching Wu // Proceedings of the IEEE-INNS-ENNS Int. Joint Conf. on Neural Networks (IJCNN 2000) "Neural Computing: New Challenges and Perspectives for the New Millennium", Como, Italy, July 27. – 2000. – Vol. 4. – P. 408–413. – DOI: 10.1109/IJCNN.2000.860806
16. Zhdanov I.A. Razrabotka sistemy avtomatizirovannogo analiza i kontrolya tekushchego sostoyaniya neftyanykh mestorozhdeniy: dis. … kand. tekhn. nauk: 2.3.1. – SPb., 2020. – 117 s.
17. Oberwinkler Ch., Stundner M. From Real-Time Data to Production Optimization // SPE Production & Facilities. – 2005. – Vol. 20, Issue 03. – P. 229–239. – DOI: 10.2118/87008-PA
18. Hochreiter S., Schmidhuber Jü. Long Short-Term Memory // Neural Computation. – 1997. – Vol. 9, Issue 8. – P. 1735–1780. – DOI: 10.1162/neco.1997.9.8.1735
19. Gol'dberg Y. Neyrosetevye metody v obrabotke estestvennogo yazyka. – M.: DMK Press, 2019. – 282 s.