Abstract
The resource potential of coalbed methane (CBM) in China is huge, the geological resources is 0.68 × l012 m3 in Chongqing upon the depth of 2000 m. The high content of coalbed methane reveals that Chongqing has a good potential of coalbed methane resources. With the artificial intelligence technology applying to many respects of unconventional oil and gas exploration and development, the neural network technology is playing a more and more important role to solve the complex geophysical model.
In this paper, the regularization dropout to long short term memory (RD-LSTM) neural network algorithm is proposed for CBM well productivity prediction. Different weights are given by the gate control unit combining with the factors that affect the production of CBM well, which is from the geology, geophysics, fracturing, and recovery technology. In order to discard the invalid operator, reduce the computational complexity of different hidden layers, regularization dropout mechanism is added to gated units such as forget gate, output gate, external output gate, et al. The regularization function is determined by the geometric average of the probability of the different unit. When the value is zero, the dropout occurs. The state of the gate would participate in the next sub-network. The result shows that the algorithm has lower complexity and higher computational efficiency than LSTM algorithm. In addition, the algorithm can accurately predict the production of the CBM well after increasing the sample data, which is beneficial to improve the recovery of CBM well, well pattern planning, process of drainage and recovery.
Copyright 2019, IFEDC Organizing Committee.
This paper was prepared for presentation at the 2019 International Field Exploration and Development Conference in Xi’an, China, 16–18 October, 2019.
This paper was selected for presentation by the IFEDC Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the IFEDC Technical Team and are subject to correction by the author(s). The material does not necessarily reflect any position of the IFEDC Technical Committee its members. Papers presented at the Conference are subject to publication review by Professional Team of IFEDC Technical Committee. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of IFEDC Organizing Committee is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of IFEDC. Contact email: paper@ifedc.org.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Liu, C.L., Zhu, J., Che, C.B., et al.: Methodologies and results of the latest assessment of coalbed methane resources in China. Nat. Gas. Ind. 29(11), 130–132 (2009)
Srinivasan, K., Ertekin, T.: Development and testing of an expert system for coalbed methane reservoirs using artificial neural networks. In: The 2008 SPE Eastern Regional/AAPG Eastern Section Joint Meeting, Pittsburgh, USA (2008)
Enyioha, C., Ertekin, T.: Advanced well structures: an artificial intelligence approach to field deployment and performance prediction. In: The SPE Intelligent Energy Conference and Exhibition, Utrecht, Netherlands (2010)
Chen, Y., Yang, Y., Luo, J.: Uncertainty analysis of coalbed methane economic assessment with montecarlo method. Procedia Environ. Sci. 12, 640–645 (2012)
Zhang, Y.Y., Sun, X.F., Shang, F.J., et al.: Study on productivity prediction and influencing factors of Qinshui coalbed methane wells. J. Petrol. Nat. Gas 34(11), 118–122 (2012)
Jiang, Y.Q., Li, C.Y., Li, Z.J.: A new gas well productivity prediction model based on fuzzy comprehensive evaluation and BP neural network. Oil Gas Field Surf. Eng. 28(10), 5–7 (2009)
Ye, S.J., Jiang, H.Q., Chen, M.F.: Productivity prediction of horizontal wells based on grey relation and neural network technology. J. Daqing Petrol. Inst. 33(3), 53–55 (2009)
Lu, Y.M., Tang, D.Z., Li, Z.P., et al.: Dynamic productivity fitting and prediction model of coalbed methane wells. J. Coal Sci. 36(9), 1481–1485 (2011)
Yang, Y.G., Qing, Y.: Stochastic dynamic model for coalbed methane productivity prediction and its application. J. Coal Sci., 122–125 (2001)
Wang, Y.: Study of methods of CBM productivity numerical simulation and its prediction. China University of Mining and Technology (2015)
Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9(8), 1735–1780 (1997)
Graves, A., Jürgen, S.: Framewise phoneme classification with bidirectional LSTM and other neural network architectures. Neural Netw. 18(5–6), 602–610 (2005)
Liu, J., Shahroudy, A., Xu, D., et al.: Spatio-temporal LSTM with trust gates for 3D human action recognition. In: Computer Vision – ECCV 2016, pp. 816–833 (2016)
Greff, K., Srivastava, R.K., Koutník, J., et al.: LSTM: a search space odyssey. IEEE Trans. Neural Netw. Learn. Syst. 28(10), 2222–2232 (2015)
Zhu, W., Lan, C., Xing, J., et al.: Co-occurrence feature learning for skeleton based action recognition using regularized deep LSTM networks. In: Computer Vision and Pattern Recognition, pp. 1–7 (2016)
Goodfellow, I., Bengio, Y., Courville, A.: Deep Learning. The MIT Press, Boston (2016)
Acknowledgments
We are grateful to be supported by Chongqing Administration of Science and Technology (No. cstc2016zdcy-zd90002, cstc2017jxjl20006, cstc2018jxjl120017), and the project “Application of Monitoring, Early Warning and Risk Control Technology in Highly Steep Bank Slope and Hydro-fluctuation Belt of Wuxia Section, Three Gorges Reservoir in Chongqing” from Chongqing Administration of Planning and Natural Resources.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Xu, H. et al. (2020). RD-LSTM Neural Network for Productivity Prediction of Coalbed Methane. In: Lin, J. (eds) Proceedings of the International Field Exploration and Development Conference 2019. IFEDC 2019. Springer Series in Geomechanics and Geoengineering. Springer, Singapore. https://doi.org/10.1007/978-981-15-2485-1_103
Download citation
DOI: https://doi.org/10.1007/978-981-15-2485-1_103
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-15-2484-4
Online ISBN: 978-981-15-2485-1
eBook Packages: EngineeringEngineering (R0)