PREDICTING THE TECHNICAL CONDITION OF AN ELECTRIC SUBMERSIBLE PUMP BASED ON NEURAL NETWORK MODELING

Igor Vladimirovich Karakulov, Andrey Vladimirovich Klyuev, Valery Yurievich Stolbov

Abstract


The problem of predicting the state of an Electric Submersible Pump during
operation is considered. Downtime and shortages caused by pump failure lead to losses in oil production and require time to replace equipment. By predicting the condition of the equipment, it is possible to minimize pump maintenance costs and reduce well downtime. Expert systems and predictive analytics methods are used to analyze the state of systems. The scientific work uses methods that are based on artificial neural networks. Purpose of research. Elaboration of the issues of forecasting the technical condition of the pump through by using machine-learning models. Materials and methods. Equipment failure forecasting is carried out using time series analysis. The data was obtained from telemetric sensors of the monitoring system installed on an electric submersible pump. The initial data were taken at one-minute intervals. Initial data preprocessing was carried out.
The data was cleared of values (peaks) that are clearly got out of normal operation and places where the phase voltage was equal to zero were removed. An artificial neural network with the LSTM
neuron type is used to predict time series. Time series forecasting was carried out for five days. Evaluating system parameters over long periods allows you to assess the condition of its components and prevent equipment failure. Results. The possibilities of neural networks trained on the basis of data from telemetric sensors of the monitoring system for predicting the values of vertical vibration of the pump are investigated. The use of a neural network model in the form of LSTM, which has shown good results in the analysis of time series, is justified. It was found that neural networks capture the trend well within the time series, which indicates the possibility of using it together with the expert system. Conclusion. The proposed methods and models are tested on real data, which confirms the possibility of their use in the development of an intelligent information system for managingthe technical condition of an Electric Submersible Pump during operation.

Keywords


Electric Submersible Pump, forecasting, time series, artificial neural network, estimation of fore-casting accuracy, LSTM network

References


Camilleri, L.A.P. How 24/7 Real-Time Surveillance Increases ESP Run Life and Uptime / L.A.P. Ca¬milleri, J. Macdonald // Society of Petroleum Engineers. – 2010, January 1. DOI: 10.2118/134702-MS

Истомин, Д.А. Экспертная система оценки технического состояния узлов электро-центробежных насосов на основе продукционного представления знаний и нечеткой ло-гики / Д.А. Истомин, В.Ю. Столбов, Д.Н. Платон // Вестник ЮУрГУ. Серия «Компью-терные технологии, управление, радиоэлектроника». – 2020. – Т. 20, № 1. – С. 133–143. DOI: 10.14529/ctcr200113

Application of a Fuzzy Expert System to Analyze and Anticipate ESP Failure Modes / D. Gras-sian, M. Bahatem, T. Scott, D. Olsen // Society of Petroleum Engineers. – 2017, November 13. DOI: 10.2118/188305-MS

Orchard, B. FuzzyCLIPS Version 6.10 d User’s Guide / B. Orchard. – National Research Council of Canada, 2004. – 82 p.

Riley, Gary. CLIPS: A Tool for Building Expert Systems [online] / Gary Riley. – GHG Corpo-ration, 2001. – http://www.ghg.net/clips/CLIPS.html (дата обращения: 19.04.2001).

Yan, Yu. Hands-On Data Science with Anaconda: Utilize the right mix of tools to create high-performance data science applications / Yu. Yan, J. Yan. – Packt Publishing-ebooks Account, 2018. – 364 p.

Брускин, С.Н. Модели и инструменты предиктивной аналитики для цифровой корпо-рации / С.Н. Брускин // Вестник РЭУ им. Г.В. Плеханова. – 2017. – № 5. – С. 136–139.

Умная платформа / А. Скобеев, Д. Маганов, В. Рогод и др. – 2019. – https://hbr-russia.ru/ innovatsii/tekhnologii/803089 (дата обращения: 11.06.2020).

Чучуева, И.А. Модель прогнозирования временных рядов по выборке максимального по-добия: дис. … канд. техн. наук / И.А. Чучуева. – М.: Московский государственный тех-нический университет им. Н.Э. Баумана, 2012.

Spatial Bayesian Network for predicting sea level rise induced coastal erosion in a small Pa-cific Island / O. Sahin, R.A. Stewart, G. Faivre et al. // Journal of Environmental Manage-ment. – May 15 2019. – Vol. 238. – P. 341–351. DOI: 10.1016/j.jenvman.2019.03.008

Bretherton, F.P. A technique for objective analysis and design of oceanographic experiments applied to MODE-73 / F.P. Bretherton, R.E. Davis, C. Fandry // Deep Sea Research and Oceanographic Abstracts. – Elsevier, 1976. – Vol. 23, no. 7. – P. 559–582.

Tealab, A. Time series forecasting using artificial neural networks methodologies: A systemat-ic review / A. Tealab // Future Computing and Informatics Journal. – 2018. – Vol. 3, no. 2. – p. 334–340.

George, E.P. Box and Gwilym M. Jenkins Time series analysis forecasting and control /

E.P. George. – San Francisco: Holden-Day, 1970. – 553 p.

Sagheer, A. Time series forecasting of petroleum production using deep LSTM recurrent net-works / A. Sagheer, M. Kotb // Neurocomputing. – 2019. – Vol. 323. – P. 203–213.

Ясницкий, Л.Н. Интеллектуальные системы / Л.Н. Ясницкий. – М.: Лаборатория зна-ний, 2016. – 221 с.

Hochreiter, S. Long short-term memory / S. Hochreiter, J. Schmidhuber // Neural Computa-tion: journal. – 1997. – Vol. 9, no. 8. – P. 1735–1780.




DOI: http://dx.doi.org/10.14529/ctcr200404

Refbacks

  • There are currently no refbacks.