Operating condition and fault diagnosis of electric submersible pump based on OCSVM

被引：0
|
作者：
Liu G. [1 ]
Du Y. [2 ]
Guo L. [1 ]
Shi E. [3 ]
Wang Z. [3 ]
Yan Z. [1 ]
机构：
[1] College of Control Science and Engineering in China University of Petroleum(East China), Qingdao
[2] College of Oceanography and Space Informatics in China University of Petroleum(East China), Qingdao
[3] Hekou Oil Production Plant, Shengli Oilfield Company, SINOPEC, Dongying
来源：
Zhongguo Shiyou Daxue Xuebao (Ziran Kexue Ban)/Journal of China University of Petroleum (Edition of Natural Science) | 2021年 / 45卷 / 05期
关键词：
Electric submersible pump; Feature extraction; One-class support vector machine(OCSVM); Operating condition and fault diagnosis;
D O I：
10.3969/j.issn.1673-5005.2021.05.019
中图分类号：
学科分类号：
摘要：
One-class support vector machine(SVM) model was used to distinguish electric submersible pump normal operation and abnormal operating states. Based on only the data in the electric submersible pump normal state, the one-class SVM model is applied to identify abnormal state data. Firstly, we preprocess the electric submersible pump current data and filter the current data under normal conditions. Then, according to the characteristics of the electric submersible pump and data characteristics, six relevant data features are extracted. The one-class SVM model is subsequently used to identify abnormal states including unknown faults, so as to realize the working conditions and fault diagnosis of the electric submersible pump. Finally, the actual production data is used to verify the model. The results prove that the method proposed in this paper has a high recognition accuracy and a strong model generalization ability. Through real-time analysis of daily operation data of the electric submersible pump, the real-time monitoring of status and early warning of abnormal working conditions of the electric submersible pump is realized. © 2021, Editorial Office of Journal of China University of Petroleum(Edition of Natural Science). All right reserved.
引用
收藏
页码：162 / 168
页数：6
相关论文
共 22 条
  • [1] SHARMAR, PANDEYN, A neural network model for electric submersible pump surveillance[C/OL], 2016 IEEE International Conference on Communication and Signal Processing, pp. 2083-2088, (2016)
  • [2] ZHAO Fangjian, Research status and development trend of fault diagnosis technology for electric submersible pump units, Petroleum Machinery, 39, 1, pp. 80-84, (2011)
  • [3] PENG Kexiang, Fault diagnosis of electric submersible pump based on BP neural network, Journal of Petrochemical College, 29, 1, pp. 76-79, (2016)
  • [4] CASTELLANOS M B, SERPA A L, BIAZUSSI J L, Et al., Fault identification using a chain of decision trees in an electrical submersible pump operating in a liquid-gas flow, Journal of Petroleum Science and Engineering, 184, 5, pp. 1-12, (2020)
  • [5] CHEN Zhiguo, FENG Ding, ZHU Muxiu, Et al., Research on current card feature extraction method based on pattern recognition, Petroleum Machinery, 32, 2, pp. 38-41, (2004)
  • [6] YAO Ken, Analysis and prediction technology of working condition of electric submersible pump, (2016)
  • [7] HAN Guoqing, CHEN Mian, ZHANG He, Et al., Real-time monitoring and diagnosis of electrical submersible pump, (2015)
  • [8] FENG Ding, WANG Peng, ZOU Yan, Et al., Research on the method of determining the center coordinates of the electric submersible pump current card, Journal of Oil and Gas Technology, 30, 1, pp. 154-156, (2008)
  • [9] LI Lu, HUA Chenquan, XU Xuan, Condition monitoring and fault diagnosis of electric submersible pump based on wellhead electrical parameters and production parameters[J], Systems Science & Control Engineering, 6, 3, pp. 253-261, (2018)
  • [10] GUPTA S, SAPUTELLIL, NIKOLAOUM, Applying big data analytics to detect, diagnose, and prevent impending failures in electric submersible pumps, (2016)
123