A MOVING WINDOW APPROACH TO IDENTIFICATION OF PATTERNS IN MULTIVARIATE TIME SERIES: APPLICATION TO THE MULTIVARIATE FAULT DETECTION OF ELECTRICAL SUBMERSIBLE PUMPS

Autor:	Sheremetov, L., Zarate-Guzmán, N.
Jazyk:	angličtina
Rok vydání:	2022
Předmět:	Multivariate time series pattern recognition reference window electrical submersible pump
DOI:	10.5281/zenodo.7038257
Popis:	The paper describes a multivariate time series pattern recognition method based on reference windows and used for the detection of fault patterns of electric submersible pumps caused by scales formed during production process in petroleum wells. Through a “moving window” strategy, the algorithm finds and selects reference windows in a long time series and computes the similarity between each selected window and the reference one (smaller time series) using the Euclidean distance. This method can simultaneously get the results of fault detection and fault diagnosis in a monitoring process. Results of algorithm analysis and fault detection experiments indicate the validity and practicability of the presented method. References: 1. Abdalla, R., Samara, H., Perozo N., Paz-Carvajal C., and Jaeger P. Machine Learning Approach for Predictive Maintenance of the Electrical Submersible Pumps (ESPs). ACS Omega 2022, 7, 21, 17641–17651, https://doi.org/10.1021/acsomega.1c05881 2. Alamu, O. A.; Pandya, D. A.; Warner, O.; Debacker, I. ESP Data Analytics: Use of Deep Autoencoders for Intelligent Surveillance of Electric Submersible Pumps. Offshore Technology Conference 2020, OTC-30468-MS. 3. Batyrshin, I., Sheremetov, L.: Time Series Pattern Recognition Based on MAP Transform and Local Trend Associations. In: J. Kittler, J. F. Martínez-Trinidad and J. A. Carrasco-Ochoa (eds.) Progress in Pattern Recognition, Image Analysis and Applications, Lecture Notes in Computer Science, Springer Verlag, Vol. 4225, pp. 910-919 (2006). 4. Bhardwaj, A. S.; Saraf, R.; Nair, G. G.; Vallabhaneni, S. Real-Time Monitoring and Predictive Failure Identification for Electrical Submersible Pumps. Abu Dhabi International Petroleum Exhibition & Conference 2019, SPE-197911-MS. 5. Feng, D., Yang, C., Tan, B., Xu, G., Yuan, Y., and Wang, P.: The Diagnosis Research of Electric Submersible Pump Based on Neural Network. In: H. Wang et al. (eds.): The Sixth ISNN 2009, AISC 56, pp. 721–727. Springer-Verlag Berlin Heidelberg (2009). 6. Fontes, C. H., and Pereira, O.: Pattern recognition in multivariate time series - A case study applied to fault detection in a gas turbine. Eng. Applications of Artificial Intelligence, 49, 10–18, (2016). http://doi.org/10.1016/j.engappai.2015.11.005 7. Jansen Van Rensburg, N.; Kamin, L.; Davis, S. Using Machine Learning-Based Predictive Models to Enable Preventative Maintenance and Prevent ESP Downtime. Abu Dhabi Int. Petroleum Exhibition & Conference 2019, SPE-197146-MS. 8. Khabibullin, R. A.; Shabonas, A. R.; Gurbatov, N. S.; Timonov, A. V. Prediction of ESPs Failure Using ML at Western Siberia Oilfields with Large Number of Wells. SPE Russian Petroleum Technology Conference 2020, SPE-201881-MS 9. Konga, F., Chen, R.: A combined method for triplex pump fault diagnosis based on wavelet transform, fuzzy logic and neuro-networks, Mech. Syst. Signal Process. 18: 161-168 (2004). 10. Kovalchuk M. S .and D. A. Poddubniy (2018) Diagnosis of Electric Submersible Centrifugal Pump. IOP Conf. Ser.: Earth Environ. Sci. 115 012026 11. Li, L.; Hua, C.; Xu, X. Condition monitoring and fault diagnosis of electric submersible pump based on wellhead electrical parameters and production parameters. Systems Science & Control Engineering 2018, 6, 253−261. 12. Liao, T.W.: Clustering of Time Series Data: A Survey. Pattern Recognition, 38(11), 1857-1874 (2005). 13. Lin, J., Keogh, E., Lonardi, S., Chiu, B.: A Symbolic Representation of Time Series, with Implications for Streaming Algorithms. In Proc. of the 8th ACM SIGMOD Workshop on Research Issues in Data Mining and Knowledge Discovery, DMKD'03, pp. 2-11. ACM, New York (2003). 14. Muralidharan, V., Sugumaran, V., Indira, V.: Fault diagnosis of monoblock centrifugal pump using SVM. Eng. Science and Technology, an Int. Journal, 17(3), 152–157 (2014). 15. Sharma, R., Pandey, N.: A Neural Network Model for Electric Submersible Pump Surveillance. In: Proc. of the Int. Conf. on Communication and Signal Processing, pp. 2083-2088. April 6-8, India (2016). 16. Yu, D., Li, Y., Sun, H., Ren, Y., Zhang, Y., and Qi, W.: A Fault Diagnosis Method for Oil Well Pump Using Radial Basis Function Neural Network Combined with Modified Genetic Algorithm, J. of Control Science and Eng., (2017), https://doi.org/10.1155/2017/5710408. 17. Yang, J.; Li, W.; Chen, J.; Sheng, L. Fault diagnosis of electric submersible pump tubing string leakage. E3S Web of Conferences 2021, 245, 01042. 18. Yuan, J. and Liu X.: Semi-supervised learning and condition fusion for fault diagnosis, Mech. Syst. Signal Process. 38: 615-627 (2013). 19. Zhang, P., Chen, T., Wang, G., Peng, C.: Ocean Economy and Fault Diagnosis of Electric Submersible Pump applied in Floating platform. Int. Journal of e-Navigation and Maritime Economy, 6 037–043 (April 2017). 20. Zhang, Q.S.: Fault tree analysis of the electric submersible pump system, Oil Field Equipment, Vol. 37, No. 4, pp.47-50 (2008). 21. Zouari, R., Sieg-Zieba, S., Sidahmed, M.: Fault Detection System for Centrifugal Pumps Using Neural Networks and Neuro-Fuzzy Techniques, In: Proc. of the Surveillance 5 Cetim Senlis, 11-13 October, (2004).
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c2ba7a424366e9db3da0147543082566 Zobrazit plný text záznamu