# Forecasting of Coalbed Methane Daily Production Based on T-LSTM Neural Networks

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Related Work

_{2.5}) concentration of air pollution and obtained more accurate predictions than by existing methods. Chen [29], Tian [30], and Li et al. [31] studied traffic flow using an LSTM model. Ma et al. [32] used LSTM and Beijing microwave traffic detection data to predict traffic speed and found that a LSTM network predicted accuracy and stability better than traditional neural network and other parametric or nonparametric algorithms did. Fischer [33] and Kim et al. [34] used the LSTM model to make predictions about financial markets and found that the results were better than the results of random forest, deep neural network (DNN), and other methods. Peng [35] and Fang et al. [36] used the LSTM model to predict electricity prices and electricity sales. Sagheer et al. [37] proposed a deep long short-term memory (DLSTM) oil production prediction method based on genetic algorithm optimization and compared it with statistical and software calculation methods, finding that the proposed DLSTM model was more accurate under different comparison standards. Wang [3] analyzed the time series of mine gas leakage based on the deep learning method but did not predict CBM production. Although LSTM models have been widely used in time series analysis, they are still relatively new in CBM production forecasting. In this study, the LSTM model was applied to predict CBM production, leading to discussion of the feasibility of using deep learning to predict CBM production.

## 3. Data and Methods

#### 3.1. Data Description

^{3}). It can be seen from the curves that the CBM output types of the seven wells are different and vary greatly with different phases.

#### 3.2. LSTM Neural Network

_{t}memory at time t. W, U, and V represent the weights of the input, the moment, and the output, respectively, and O represents the output values. RNN can capture timing features but can solve only short-sequence problems; it does not work for long-sequence data such as CBM daily production.

_{t}and h

_{t}represent the input and output at time t, respectively; and h

_{t–1}represents the output at the previous time. The LSTM network structure solves the RNN gradient problem through three “gates”: a “forget gate”, an “input gate”, and an “output gate”. The “forget gate” decides which information to discard and which to retain, the “input gate” controls the information input to the cell, and the “output gate” controls the output information. The hidden layer structure of the LSTM network can be calculated by the following equations:

_{i}is predicted CBM production, A

_{i}is actual CBM production, and n is number of test samples.

#### 3.3. T-LSTM Model

## 4. T-LSTM Network Training and Parameter Optimization

## 5. Results and Discussion

## 6. Conclusions

## Author Contributions

## Funding

## Conflicts of Interest

## References

- Yun, M.G.; Rim, M.W.; Han, C.N. A model for pseudo-steady and non-equilibrium sorption in coalbed methane reservoir simulation and its application. J. Nat. Gas Sci. Eng.
**2018**, 54, 342–348. [Google Scholar] [CrossRef] - Xu, H. Research on Particle Swarm Optimization Algorithm Improvement and Its Application in the CBM Production Forecast; China University of Mining and Technology: Jiangsu, China, 2013. [Google Scholar]
- Wang, S.Q. Gas Time Series Prediction and Anomaly Detection Based on Deep Learning; China University of Mining and Technology: Jiangsu, China, 2018. [Google Scholar]
- Ping, X.; Xiaochun, L.; Zhiming, F.; Bing, B. Application of Quantification Theory to predict Coal Methane Content. Disaster Adv.
**2012**, 5, 1609–1614. [Google Scholar] - Clarkson, C. Production data analysis of unconventional gas wells: Review of theory and best practices. Int. J. Coal Geol.
**2013**, 109, 101–146. [Google Scholar] [CrossRef] - Li, H.; Li, Z.; Zhang, H.; Hu, J.T. Study of Coalbed Methane Production Forecast at Different Stages by Using Weibull Model. J. Oil Gas Technol.
**2013**, 35, 100–103. [Google Scholar] - Xu, B.X.; Li, X.F.; Hu, X.H.; Hu, S.M. Type curves for production prediction of coalbed methane wells. J. China Univ. Min. Technol.
**2011**, 40, 743–747. [Google Scholar] - Jang, H.; Kim, Y.; Park, J.; Lee, J. Prediction of production performance by comprehensive methodology for hydraulically fractured well in coalbed methane reservoirs. Int. J. Oil Gas Coal Technol.
**2019**, 20, 143–168. [Google Scholar] [CrossRef] - Chen, X.J.; Li, P.F.; Li, P.; Hui, P.; Guo, Y.H. Application of Multiple Stepwise Regression Analysis in Prediction of Coal Seam Gas Content (in Chinese). Coal Eng.
**2019**, 51, 106–111. [Google Scholar] - Li, D.; Cheng, S.; Wang, Z. The Prediction on Coal Field’s CBM (Coalbed Methane) Resource (in Chinese). Shanxi Sci. Technol.
**2015**, 1, 54–56. [Google Scholar] - Cipolla, C.L.; Lolon, E.P.; Erdle, J.C.; Rubin, B. Reservoir Modeling in Shale-Gas Reservoirs. SPE Reserv. Eval. Eng.
**2010**, 13, 638–653. [Google Scholar] [CrossRef] - Zhao, Y.-L.; Zhao, L.; Wang, Z.-M.; Yang, H. Numerical simulation of multi-seam coalbed methane production using a gray lattice Boltzmann method. J. Pet. Sci. Eng.
**2019**, 175, 587–594. [Google Scholar] [CrossRef] - Zhou, F. History matching and production prediction of a horizontal coalbed methane well. J. Pet. Sci. Eng.
**2012**, 96, 22–36. [Google Scholar] [CrossRef] - Lü, Y.; Tang, D.; Xu, H.; Tao, S. Productivity matching and quantitative prediction of coalbed methane wells based on BP neural network. Sci. China Ser. E Technol. Sci.
**2011**, 54, 1281–1286. [Google Scholar] [CrossRef] - King, G. Material-Balance Techniques for Coal-Seam and Devonian Shale Gas Reservoirs with Limited Water Influx. SPE Reserv. Eng.
**1993**, 8, 67–72. [Google Scholar] [CrossRef] - Shi, J.; Chang, Y.; Wu, S.; Xiong, X.; Liu, C.; Feng, K. Development of material balance equations for coalbed methane reservoirs considering dewatering process, gas solubility, pore compressibility and matrix shrinkage. Int. J. Coal Geol.
**2018**, 195, 200–216. [Google Scholar] [CrossRef] - Sun, Z.; Shi, J.; Zhang, T.; Wu, K.; Miao, Y.; Feng, D.; Sun, F.; Han, S.; Wang, S.; Hou, C.; et al. The modified gas-water two phase version flowing material balance equation for low permeability CBM reservoirs. J. Pet. Sci. Eng.
**2018**, 165, 726–735. [Google Scholar] [CrossRef] - Liu, Z.; Yang, X.; Zhang, J. Logging Predicting for Coalbed Gas Content in Eastern Block of Ordos Basin. Geol. Sci. Technol. Inf.
**2014**, 33, 95–99. [Google Scholar] - Xu, H.; Tang, D.Z.; Liu, D.M.; Tang, S.H.; Yang, F.; Chen, X.Z.; He, W.; Deng, C.M. Study on coalbed methane well productivity by using artificial neural network (in Chinese). China Coal
**2012**, 38, 9–13. [Google Scholar] - Lv, Y.M.; Tang, D.Z.; Li, Z.P.; Shao, X.J.; Hu, H. Fitting and predicting models for coalbed methane wells dynamic productivity (in Chinese). J. China Coal Soc.
**2011**, 36, 1481–1485. [Google Scholar] - Ma, X.; Zhang, C.; Chen, X.F. A Method Combined Principal Component Analysis and BP Artifical Neural Network for Coalbed Methane (CBM) Wells to Predict Productivity (in Chinese). Sci. Technol. Ind.
**2013**, 13, 97–100. [Google Scholar] - Xia, H.; Qin, Y.; Zhang, L.; Cao, Y.; Xu, J. Forecasting of coalbed methane (CBM) productivity based on rough set and least squares support vector machine. In Proceedings of the 2017 25th International Conference on Geoinformatics, Buffalo, NY, USA, 2–4 August 2017; pp. 1–6. [Google Scholar]
- Li, C.S.; Tan, M.X.; Zhang, K.J. Research on Single Well Production Prediction Based on Improved BP Neural Networks (in Chinese). Sci. Technol. Eng.
**2011**, 11, 7766–7769. [Google Scholar] - Yang, Y.G.; Qin, Y. Study and application on random dynamic model of the coalbed methane output forecasting (in Chinese). J. China Coal Soc.
**2001**, 2, 122–125. [Google Scholar] - Bai, Y.; Zeng, B.; Li, C.; Zhang, J. An ensemble long short-term memory neural network for hourly PM2.5 concentration forecasting. Chemosphere
**2019**, 222, 286–294. [Google Scholar] [CrossRef] [PubMed] - Li, X.; Peng, L.; Yao, X.; Cui, S.; Hu, Y.; You, C.; Chi, T. Long short-term memory neural network for air pollutant concentration predictions: Method development and evaluation. Environ. Pollut.
**2017**, 231, 997–1004. [Google Scholar] [CrossRef] [PubMed] - Wen, C.; Liu, S.; Yao, X.; Peng, L.; Li, X.; Hu, Y.; Chi, T. A novel spatiotemporal convolutional long short-term neural network for air pollution prediction. Sci. Total. Environ.
**2019**, 654, 1091–1099. [Google Scholar] [CrossRef] - Zhao, J.; Deng, F.; Cai, Y.; Chen, J. Long short-term memory - Fully connected (LSTM-FC) neural network for PM2.5 concentration prediction. Chemosphere
**2019**, 220, 486–492. [Google Scholar] [CrossRef] - Zhao, Z.; Chen, W.; Wu, X.; Chen, P.C.Y.; Liu, J. LSTM network: A deep learning approach for short-term traffic forecast. IET Intell. Transp. Syst.
**2017**, 11, 68–75. [Google Scholar] [CrossRef] [Green Version] - Tian, Y.; Zhang, K.; Li, J.; Lin, X.; Yang, B. LSTM-based traffic flow prediction with missing data. Neurocomputing
**2018**, 318, 297–305. [Google Scholar] [CrossRef] - Li, Y.; Cao, H. Prediction for Tourism Flow based on LSTM Neural Network. Procedia Comput. Sci.
**2018**, 129, 277–283. [Google Scholar] [CrossRef] - Ma, X.; Tao, Z.; Wang, Y.; Yu, H.; Wang, Y. Long short-term memory neural network for traffic speed prediction using remote microwave sensor data. Transp. Res. Part C Emerg. Technol.
**2015**, 54, 187–197. [Google Scholar] [CrossRef] - Fischer, T.; Krauss, C. Deep learning with long short-term memory networks for financial market predictions. Eur. J. Oper. Res.
**2018**, 270, 654–669. [Google Scholar] [CrossRef] [Green Version] - Kim, H.Y.; Won, C.H. Forecasting the volatility of stock price index: A hybrid model integrating LSTM with multiple GARCH-type models. Expert Syst. Appl.
**2018**, 103, 25–37. [Google Scholar] [CrossRef] - Peng, L.; Liu, S.; Liu, R.; Wang, L. Effective long short-term memory with differential evolution algorithm for electricity price prediction. Energy
**2018**, 162, 1301–1314. [Google Scholar] [CrossRef] - Fang, Z.Q.; Wang, X.H.; Xia, T. Electricity Sales Forecasting Based on Long-short Term Memory Networks. Electr. Power Eng. Technol.
**2018**, 37, 78–83. (In Chinese) [Google Scholar] - Sagheer, A.; Kotb, M. Time series forecasting of petroleum production using deep LSTM recurrent networks. Neurocomputing
**2019**, 323, 203–213. [Google Scholar] [CrossRef] - Chen, Z.; Mauricio, A.; Li, W.; Gryllias, K. A deep learning method for bearing fault diagnosis based on Cyclic Spectral Coherence and Convolutional Neural Networks. Mech. Syst. Signal Process.
**2020**, 140, 1–16. [Google Scholar] [CrossRef] - Chemali, E.; Kollmeyer, P.; Preindl, M.; Ahmed, R.; Emadi, A.; Kollmeyer, P. Long Short-Term Memory Networks for Accurate State-of-Charge Estimation of Li-ion Batteries. IEEE Trans. Ind. Electron.
**2017**, 65, 6730–6739. [Google Scholar] [CrossRef] - Cortez, B.; Carrera, B.; Kim, Y.-J.; Jung, J.-Y. An architecture for emergency event prediction using LSTM recurrent neural networks. Expert Syst. Appl.
**2018**, 97, 315–324. [Google Scholar] [CrossRef] - Kumar, J.; Goomer, R.; Singh, A.K. Long Short Term Memory Recurrent Neural Network (LSTM-RNN) Based Workload Forecasting Model For Cloud Datacenters. Procedia Comput. Sci.
**2018**, 125, 676–682. [Google Scholar] [CrossRef] - Li, C.; Wang, Z.; Rao, M.; Belkin, D.; Song, W.; Jiang, H.; Yan, P.; Li, Y.; Lin, P.; Hu, M.; et al. Long short-term memory networks in memristor crossbar arrays. Nat. Mach. Intell.
**2019**, 1, 49–57. [Google Scholar] [CrossRef] - Petersen, N.C.; Rodrigues, F.; Pereira, F.C. Multi-output bus travel time prediction with convolutional LSTM neural network. Expert Syst. Appl.
**2019**, 120, 426–435. [Google Scholar] [CrossRef] [Green Version] - Gao, W.; Farahani, M.R.; Aslam, A.; Hosamani, S. Distance learning techniques for ontology similarity measuring and ontology mapping. Clust. Comput.
**2017**, 20, 959–968. [Google Scholar] [CrossRef] - Xiong, Z.; Wu, Y.; Ye, C.; Zhang, X.; Xu, F. Color image chaos encryption algorithm combining CRC and nine palace map. Multimedia Tools Appl.
**2019**, 78, 31035–31055. [Google Scholar] [CrossRef] - Cordero, A.; Jaiswal, J.P.; Torregrosa, J.R. Stability analysis of fourth-order iterative method for finding multiple roots of non-linear equations. Appl. Math. Nonlinear Sci.
**2019**, 4, 43–56. [Google Scholar] [CrossRef] [Green Version] - Voit, M.; Meyer-Ortmanns, H. Predicting the separation of time scales in a heteroclinic network. Appl. Math. Nonlinear Sci.
**2019**, 4, 279–288. [Google Scholar] [CrossRef] [Green Version] - Bergmeir, C.; Benítez, J.M. On the use of cross-validation for time series predictor evaluation. Inf. Sci.
**2012**, 191, 192–213. [Google Scholar] [CrossRef]

**Figure 4.**Framework of time series forecasting of CBM production based on LSTM. A recursive arrow indicates that the processing can be repeated.

**Table 1.**Root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) values of the seven wells.

Well | RMSE (m^{3}) | MAE (m^{3}) | MAPE (%) |
---|---|---|---|

1 | 155.04 | 149.34 | 5.07 |

2 | 25.79 | 20.17 | 0.83 |

3 | 90.86 | 74.68 | 1.37 |

4 | 47.33 | 27.14 | 1.97 |

5 | 89.02 | 64.55 | 1.67 |

6 | 174.53 | 151.75 | 2.16 |

7 | 184.06 | 148.71 | 2.59 |

**Table 2.**Comparison of the average relative error between the Transfer-LSTM (T-LSTM) model and some traditional cases in the literature.

Prediction Model | Average Relative Error (%) |
---|---|

BP neural networks [2] | 6.04 |

SVR [2] | 4.28 |

HPSO-SVR [2] | 2.44 |

IPSO-SVM [2] | 2.44 |

HPSO-SVM [2] | 2.20 |

Type curves [7] | 16 |

Decline curves [8] | 5 |

Multiple stepwise regression [9] | 13.6 |

Multiple regression [13] | 7.87 |

BP neural networks [13] | 2.25 |

BP neural networks [15] | 1.35 |

BP neural networks [16] | 4.61 |

LS-SVM [18] | 7.91 |

T-LSTM | 2.20 |

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Xu, X.; Rui, X.; Fan, Y.; Yu, T.; Ju, Y.
Forecasting of Coalbed Methane Daily Production Based on T-LSTM Neural Networks. *Symmetry* **2020**, *12*, 861.
https://doi.org/10.3390/sym12050861

**AMA Style**

Xu X, Rui X, Fan Y, Yu T, Ju Y.
Forecasting of Coalbed Methane Daily Production Based on T-LSTM Neural Networks. *Symmetry*. 2020; 12(5):861.
https://doi.org/10.3390/sym12050861

**Chicago/Turabian Style**

Xu, Xijie, Xiaoping Rui, Yonglei Fan, Tian Yu, and Yiwen Ju.
2020. "Forecasting of Coalbed Methane Daily Production Based on T-LSTM Neural Networks" *Symmetry* 12, no. 5: 861.
https://doi.org/10.3390/sym12050861