2016-Sustainable Industrial Processing Summit
SIPS 2016 Volume 9: Molten Salts and Ionic Liquids, Energy Production

Editors:Kongoli F, Gaune-Escard M, Turna T, Mauntz M, Dodds H.L.
Publisher:Flogen Star OUTREACH
Publication Year:2016
Pages:390 pages
ISSN:2291-1227 (Metals and Materials Processing in a Clean Environment Series)
CD shopping page

    Modeling method based on iterative UKFNN pumping oil production process

    Li Taifu1; Xiao-Dong Liang2; Zhou Pan3; Tang Haihong1;
    Type of Paper: Regular
    Id Paper: 272
    Topic: 17


    It is difficult to use the static modeling methods to describe pumping machine mining process because of multi-variable nonlinear and time-varying characteristics. This paper proposes a new modeling approach based on Iterated Unscented Kalman Filter Neural Networks. Firstly the algorithm uses the input data to predict state variables and the covariance matrix. Secondly using the previous estimate data to resample sigma points and do unscented transforming in order to obtain the latest sampling points. Lastly, the machine mining process model with a good precision is obtained by updating state. After doing an experiment on actual production data of a certain oilfield, the results show that the proposed method in this paper has a higher modeling precision and the stronger generalization ability and stronger real-time tracking ability than UKFNN modeling method, proposed approach is an approach choice for pumping machine mining process.


    Engineering; Petroleum; Production;


    [1] W Qing, L Wenyan, X Hao. Study on energy saving for nodding donkey oil pump, Artificial Intelligence, Management Science and Electronic Commerce (AIMSEC), 2011 2nd International Conference on. IEEE, 2011:4362-4364.
    [2] M Wenzhong, Z Hongmei, S Jinyan: Research on the Design of Linear Drive Motor of New Pumping Unit, Advanced Materials Research, 43(2012), 433-440
    [3] L Yuan, W Haitao, L Hongbin. Second International Conference on Digital Manufacturing and Automation, The Kinematic Analysis and Parametric Design of Beam Pumping Unit Based on ADAMS. IEEE, 2011:1296-1300
    [4] S Rajasekaran: Analysis of curved beams using a new differential transformation based curved beam element, Meccanica, 49(2014), 863-886
    [5] W Qing Yu, Z Hai Quan: Research and application of intelligent air pumping control technology for pumping unit, Petroleum Instrument, 25(2011), 60-62
    [6] S Qihong, R Xinhua, S Zhigang: A Design Y/Δ Energy Saving Controller for Oil-pumping Units Based on ATT7022A, Science technology and Engineering, 12(2012), 4287-4292
    [7] W Cai Yun, L Ying, L Yuanchun: Neural network modeling and optimization algorithm design for oil production control system, Journal of Jilin University, 24(2006), 164-170
    [8] D Bao, T Hai-Yan, Wei-Gui Q. Research on FNN Energy Saving Control for Light Load Oil Well with Intermittent Oil Extraction, Communications, Circuits and Systems Proceedings, 2006 International Conference on. IEEE, 2006:2034 - 2037
    [9] T Jing Wen. Research on Intelligent Optimization control Method for Oil Pumping, Adances in Mechanical Engineering, 2014
    [10] L Xingping, X Junpeng. The Method of Energy Saving in Beam Pumping Unit Based on Genetic Algorithm. Aasri Procedia, 1(2012):441-447.
    [11] Corporation H P. Research on Intelligent Optimization Control Method for Oil Pumping[J]. Advances in Mechanical Engineering, 2015, 6(8):926958-926958.
    [12] G Xiao Hua,L Tai Fu, Evolution Modeling on Beam Pumping Process by Unscent- ed Kalman Filter Neural Networks, 2014,ICRSM
    [13] L Tai Fu, H Jie, Y Jun: Nonlinear dynamic process modeling based on subspace approximation of UKF neural network. Journal of Applied Science and Engineering, 21(2013):185-194

    Full Text:

    Click here to access the Full Text

    Cite this article as:

    Taifu L, Liang X, Pan Z, Haihong T. Modeling method based on iterative UKFNN pumping oil production process. In: Kongoli F, Gaune-Escard M, Turna T, Mauntz M, Dodds H.L., editors. Sustainable Industrial Processing Summit SIPS 2016 Volume 9: Molten Salts and Ionic Liquids, Energy Production. Volume 9. Montreal(Canada): FLOGEN Star Outreach. 2016. p. 223-232.