Traditional methane extraction concentration prediction methods have limited capability in capturing complex temporal features and responding to dynamic operating conditions. To address these limitations, a multi-parameter dynamic prediction model for methane extraction concentration based on the long short-term memory (LSTM) neural network was developed. The LSTM-based prediction model was implemented in a Python environment. Monitoring data collected from a key monitoring point on the main pipeline of a methane extraction system in a mine located in Jiaozuo mining area were used as the sample dataset. Data preprocessing, including missing value imputation, outlier correction, and normalization, was performed to improve data quality. Model hyperparameters, including the time step, number of neurons, and batch size, were optimized using an early stopping strategy during the training process. The results show that the optimized LSTM model exhibits high prediction accuracy and good trend-fitting performance. The mean absolute error (MAE) is 0.585%, the root mean square error (RMSE) is 0.759%, and the mean absolute percentage error (MAPE) is 2.90%. The proposed model effectively predicts the temporal variation of methane extraction concentration with low prediction error and stable overall performance. The findings provide a theoretical basis for intelligent regulation of coal mine methane extraction systems and offer technical support for efficient methane extraction.
WEN Zhi-hui, JIA Ran, WEI Jian-ping, PAN Li-ping.
Construction and Application of Multi-parameter Prediction Model for Gas Extraction Concentration Based on LSTM[J].
Science Technology and Engineering, 2026, 26(25): 10751-10759 DOI:10.12404/j.issn.1671-1815.2507279
鉴于瓦斯抽采浓度数据呈现出非平稳性、时序性强、波动幅度大等特点,基于LSTM提出了多参数瓦斯抽采浓度预测模型,以提高预测精度与模型稳定性。预处理后的瓦斯抽采浓度、抽采纯流量、抽采负压多变量数据为输入变量,结合不同时间步长滑动窗口构建时间序列特征矩阵,通过初步搭建基础模型结构,确定网络结构中隐藏层的层数和每层神经元的数量;在LSTM层引用tanh激活函数和L2正则化约束,抑制过拟合现象。其次在全连接层引用线性整流函数(rectified linear unit,ReLU)减少梯度消失。
ZhaoWei, DongHuzi, YanZhida, et al. Hierarchical occurrence law of gas content in deep coal seams and itsrelationship with outburst prevention[J]. Journal of China Coal Society, 2025, 50(3): 1555-1568.
[3]
WangK, DuF. Coal-gas compound dynamic disasters in China: a review[J]. Process Safety and Environmental Protection, 2020(133): 1-17.
ZhangHuijie, HuoZhonggang, ShuLongyong, et al. Research status and prospect of intelligent management and control of coal mine gas extraction system[J]. Coal Science and Technology, 2025, 53(S2): 103-115.
WangGuofa, PangYihui, RenHuaiwei, et al. System engineering and key technologies research and practice of smart mine[J]. Journal of China Coal Society, 2024, 49(1): 181-202.
YangBaolin, LiuJie, LouLaizong. Prediction and early warning method for gas overrun based on PSO optimized BP neural network and SVM[J]. Science Technology and Engineering, 2025, 25(26): 11098-11106.
LiShugang, MaLi, PanShaobo, et al. Research on prediction model of gas concentration based on RNN in coal mining face[J]. Coal Science and Technology, 2020, 48(1): 33-38.
HuQingsong, ZhengShuo, LiShiyin, et al. An intelligent prediction method of gas concentration in coal mines based on improved TCN-TimeGAN[J]. Coal Science and Technology, 2024, 52(S2): 321-330.
LiangYunpei, LiShang, LiQuangui, et al. Prediction of gas concentration in the upper corner of mining working face based onthe FEDformer-LGBM-AT architecture[J]. Journal of China Coal Society, 2025, 50(1): 360-378.
WanYu, QiJinping, ZhangRu, et al. Risk prediction of coal mine gas explosion based on information gain-support vector machine (IG-SVM) model[J]. Science Technology and Engineering, 2021, 21(9): 3544-3549.
WanYu, QiJinping, ZhangRu, et al. Prediction of coal and gas outburst based on over-sampling support vector machine[J]. Science Technology and Engineering, 2021, 21(28): 12080-12087.
LiuChao, LeiChen, LiShugang, et al. Prediction model of gas concentration based on CNN-GRU and its application[J]. Journal of Safety Science and Technology, 2022, 18(9): 62-68.
[24]
WangQ, ChenH, XuN, et al. Research on the spatial-temporal series prediction method of methane concentration based on GCN-Crossformer modeling[J]. Measurement, 2025, 256: 118217.
HuangKaibo, ZhuQuanjie, ZhangErhui. Based on grey theory and BP neural network gas emission prediction research[J]. Journal of North China Institute of Science and Technology, 2020, 17(2): 16-22.
[27]
LinH F, LiW J, LiS G, et al. Coal mine gas emission prediction based on multifactor time series method[J]. Reliability Engineering & System Safety, 2024, 252: 14.
[28]
TibshiraniR. Regression shrinkage and selection via the lasso[J]. Journal of the Royal Statistical Society Series B: Statistical Methodology, 1996, 58(1): 267-288.
LiShanmei, WangDuanyang, TangRui, et al. Prediction of airport arrival delay level based on spatiotemporal association rules and LSTM[J]. China Safety Science Journal, 2025, 35(4): 59-66.
ZhuZhaokun, LiJinbao. Multi-Feature information fusion LSTM-RNN detection for OSA[J]. Journal of Computer Research and Development, 2020, 57(12): 2547-2555.
ZhangZitong, ShiChenyi, FuYingkun, et al. Production forecast of shale gas wells based on long short-term memory and Bayesian optimization[J]. Science Technology and Engineering, 2025, 25(27): 11569-11580.
[37]
郑豫. 基于深度学习的瓦斯浓度预测及安全管理研究[D]. 淮南: 安徽理工大学, 2024.
[38]
ZhengYu. Research on gas concentration prediction and safetymanagement based on deep learning[D]. Huainan: Anhui University of Science and Technology, 2024.
[39]
PreethiD, KhareN. EFS-LSTM (ensemble-based feature selection with LSTM) classifier for intrusion detection system[J]. International Journal of e-Collaboration (IJeC), 2020, 16(4): 72-86.
[40]
BarzegarR, AalamiM T, AdamowskiJ. Short-term water quality variable prediction using a hybrid CNN-LSTM deep learning model[J]. Stochastic Environmental Research and Risk Assessment, 2020, 34(2): 415-433.
LiuYongfu, ZhangTianying, HuoDianyang, et al. Simultaneous prediction for multiple load using singular spectrum analysis and BiLSTM[J]. Science Technology and Engineering, 2025, 25(19): 8099-8107.
ZhangYuanmeng, LiShaobo, ZhouPeng, et al. Residual network deep learning recommendation model based on multi-head attention mechanism[J]. Computer & Digital Engineering, 2024, 52(7): 1955-1958.
YuLi, LiHuiyuan, JiaoChenlu, et al. Trajectory prediction in complex scenes based on multi-head attention adversarial mechanism[J]. Chinese Journal of Computers, 2022, 45(6): 1133-1146.
[47]
HasegawaK, FukamiK, MurataT, et al. Machine-learning-based reduced-order modeling for unsteady flows around bluff bodies of various shapes[J]. Theoretical and Computational Fluid Dynamics, 2020, 34(4): 367-383.