Accurate prediction of coal mine gas emission is of great significance for ensuring safe production in coal mines. However, traditional prediction methods often fail to fully capture the complex nonlinear relationships between gas emissions and multidimensional features. To address the issues of data insufficiency and improve model accuracy, a gas emission prediction method based on generative adversarial networks (GAN) and a dual-parallel spatiotemporal convolutional attention architecture was proposed. First, synthetic data with a distribution similar to that of the source data were generated through GAN to augment the training dataset and enhance the robustness of the model. Subsequently, a spatiotemporal convolutional network was employed to extract spatial dependency features from the data, and a temporal attention mechanism was integrated to focus on features at critical time steps, thereby further strengthening the model's capability in modeling spatiotemporal nonlinear relationships. Through a comparison of 30 mainstream machine learning algorithms, XGBoost was selected as the core model and combined with the spatiotemporal convolutional attention feature extraction module. Experimental results indicate that the proposed model effectively improves prediction accuracy and stability across different mining areas and under multi-source heterogeneous data conditions. Compared with traditional regression models and classical deep learning methods, the model demonstrates superior performance in terms of the coefficient of determination (R2) and mean absolute error (MAE), among other metrics.
FU Tian-yu, CHEN Qiao-jun, HU Man-gu, ZHANG Xiao-jun, YANG Xiao-bin, LÜ Fu.
Novel Gas Emission Prediction Model Based on Dual-parallel Spatiotemporal Convolutional Attention in Small Sample Data[J].
Science Technology and Engineering, 2026, 26(29): 12507-12520 DOI:10.12404/j.issn.1671-1815.2600356
针对数据不足的问题,近年来各领域涌现出多种数据生成算法,如应用生成对抗网络(generative adversarial networks,GAN)提升医学图像分割的质量、大型语言模型(large language models,LLMs)来生成文本数据,以及可用于模拟高维离散变量分布的GAN[19-20]。这些方法所合成的虚拟数据与真实数据在分布特征上相当接近,为医学、计算机视觉和自然语言处理等众多领域提供了新的解决方案。由此推及瓦斯涌出量预测,也显示出极大的应用潜力。
WangL, SunY, ZhengS, et al. How efficient coal mine methane control can benefit carbon-neutral target: evidence from China[J]. Journal of Cleaner Production, 2023, 424: 138895.
CaoJie, HuQianting, ZhaoXusheng, et al. Experimental study on the influence of initial desorbed gas from crushed coal on the dynamic effects of coal and gas outburst[J]. Journal of China Coal Society, 2025, 50(9): 4260-4276.
[4]
ChenQ, QuH, LiuC, et al. Spontaneous coal combustion temperature prediction based on an improved grey wolf optimizer-gated recurrent unit model[J]. Energy, 2025, 314: 133980.
YangXiaobin, HanChao, HuMangu, et al. A spatiotemporal feature-based prediction model for gas emission: an interpretability analysis[J]. Journal of Safety and Environment, 2025, 25(10): 3794-3802.
YangBaolin, LiuJie, LouLaizong, et al. 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.
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.
[11]
KaracanC Ö, OleaR A, GoodmanG. Geostatistical modeling of the gas emission zone and its in-place gas content for Pittsburgh-seam mines using sequential Gaussian simulation[J]. International Journal of Coal Geology, 2012, 90: 50-71.
[12]
RoyA, ChoiY. Temperature dependence of source specific volatility basis sets for motor vehicle exhaust[J]. Atmospheric Environment, 2015, 119: 258-261.
[13]
YakovlevV V, KaluginY I, KaluginA Y, et al. Mathematical modelling of enhanced gas production[J]. Journal of Petroleum Exploration and Production Technology, 2019, 9(1): 561-571.
ChenQiaojun, YuHao, LiYanchang, et al. Prediction of gas emission quantity in mining face based on KPCA-LSSVM[J]. Journal of Safety Science and Technology, 2024, 20(4): 78-84.
WangQin. Research on coal mine gas concentration prediction method based on spatiotemporal deep learning and nonparametric estimation[D]. Huainan: AnHui University of Science and Technology, 2025.
XuYaosong, BaiJining, WangYuhong, et al. Prediction model of gas emission based on CEEMDAN-DA-GRU[J]. Chinese Journal of Sensors and Actuators, 2023, 36(3): 441-448.
ChenGuangsheng. Research and application of IPCA-CPSO-BP neural network model for gas emission prediction[D]. Jiaozuo: Henan Polytechnic University, 2024.
ChenYunfei. Research on dynamic prediction of gas emission in mining face based on KPCA-GA-XGBoost model[D]. Xuzhou: China University of Mining and Technology, 2024.
[24]
ChengL, LiL, LiS, et al. Prediction of gas concentration evolution with evolutionary attention-based temporal graph convolutional network[J]. Expert Systems with Applications, 2022, 200: 116944.
[25]
XuN, WangX, MengX, et al. Gas concentration prediction based on IWOA-LSTM-CEEMDAN residual correction model[J]. Sensors, 2022, 22(12): 4412.
[26]
LinH, LiW, LiS, et al. Coal mine gas emission prediction based on multifactor time series method[J]. Reliability Engineering & System Safety, 2024, 252: 110443.
[27]
WuB, ZhangW, GuoZ, et al. Research on forecasting model of gas emission in coal mining and heading face based on ARIMA-GM method[C]// Machinery, Materials Science and Energy Engineering (ICMMSEE 2015) Proceedings of the 3rd International Conference. Singapore: World Scientific Publishing Co. Pte. Ltd., 2015: 685-692.
[28]
ThambawitaV, SalehiP, SheshkalS A, et al. SinGAN-Seg: synthetic training data generation for medical image segmentation[J]. PloS ONE, 2022, 17(5): e0267976.
[29]
ChuiK T, LeeL K, WangF L, et al. A review of data augmentation and data generation using artificial intelligence in education[C]// International Conference on Technology in Education. Singapore: Springer Nature Singapore, 2023: 242-253.
[30]
WuX, HanY, ZhangM, et al. GAN-based pseudo random number generation optimized through genetic algorithms[J]. Complex & Intelligent Systems, 2025, 11(1): 31.
FengZhicong, ZhuYun, GaoFeng. Steam volume prediction of thermal power plant based on BLS-Lasso combined model[J]. Science Technology and Engineering, 2022, 22(26): 11394-11401.
ChenCong, HouLei, LiLele, et al. Prediction of aircraft fuel flow based on recurrent neural network[J]. Science Technology and Engineering, 2021, 21(27): 11663-11673.
[35]
YunitaA, PratamaM H D I, AlmuzakkiM Z, et al. Performance analysis of neural network architectures for time series forecasting: a comparative study of RNN, LSTM, GRU, and hybrid models[J]. MethodsX, 2025, 15: 103462.
[36]
GongJ, QuZ, ZhuZ, et al. Ensemble models of TCN-LSTM-LightGBM based on ensemble learning methods for short-term electrical load forecasting[J]. Energy, 2025, 318: 134757.
[37]
FanZ, GouJ, WengS. Complementary CatBoost based on residual error for student performance prediction[J]. Pattern Recognition, 2025, 161: 111265.
BiLingyue, WangQiang, WuYuhang. DTS synthesis gas pipeline leakage or damage detection and location based on adaptive variance threshold algorithm[J]. Science Technology and Engineering, 2025, 25(12): 5037-5044.
[40]
NeubauerA, BrandtS, KriegelM. Explainable multi-step heating load forecasting: using SHAP values and temporal attention mechanisms for enhanced interpretability[J]. Energy and AI, 2025, 20: 100480.
XieLirong, WangJinrui, MutellipAhmat, et al. Prediction for coal mine gas emission based on LVQ-GA-BP neural network[J]. Safety in Coal Mines, 2017, 48(12): 150-152, 156.
LiuPeng, WeiHuizi, JingJiangbo, et al. Predicting technology of gas emission quantity in coal mine based on enhanced CART regression algorithm[J]. Coal Science and Technology, 2019, 47(11): 116-122.
ZhangKexue, LiZhongxu, ChenXuexi, et al. Gas emission prediction of intelligent mines based on PCA-HPO-ELM[J]. Journal of Mining Science and Technology, 2025, 10(5): 879-889.
SunJianhua, ZhangZhili, ShiQian, et al. Prediction of gas emission based on principal component-stepwise regression analysis[J]. Coal Engineering, 2020, 52(1): 89-94.
WangLei, LiuYu, LiuZhizhong, et al. Research on prediction model for gas emission based on IABC-LSSVM[J]. Transducer and Microsystem Technologies, 2022, 41(2): 34-38.
ZhouXihua, SunJiazheng. Prediction of gas emission based on principal factor analysis and improved BP neural network[J]. Mining Safety & Environmental Protection, 2018, 45(6): 43-47, 52.
WuFengliang, HuoYuan, GaoJianan. Coal mine gas emission prediction method based on random forest regression[J]. Industry and Mine Automation, 2021, 47(8): 102-107.
LiWenhua, YangZining, WangLaigui. Application of gas emission dynamic prediction based on coupled neural network[J]. Journal of Liaoning Technical University(Natural Science), 2018, 37(4): 750-755.
WangYuanbin, LiYuanyuan, HanQian, et al. Gas emission prediction of the stope in coal mine based on PCA-BO-XGBoost[J]. Journal of Xi'an University of Science and Technology, 2022, 42(2): 371-379.
YuJian, SuiLili, CangDingbang, et al. Gas emission quantity prediction model based on kernel entropy component analysis and support vector machine combination algorithm[J]. Mathematics in Practice and Theory, 2020, 50(23): 93-99.
XuGang, WangLei, JinHongwei, et al. Gas emission prediction in mining face by factor analysis and BP neural network coupling model[J]. Journal of Xi'an University of Science and Technology, 2019, 39(6): 965-971.
FengShengcheng, ShaoLiangshan, LuWanjie, et al. Application of PCA-PSO-LSSVM model in gas emission prediction[J]. Journal of Liaoning Technical University(Natural Science), 2019, 38(2): 124-129.
XuYaosong, QiCuiyu, FengShengcheng. Gas emission prediction model based on IGSA-BP network[J]. Journal of Electronic Measurement and Instrumentation, 2019, 33(5): 111-117.
MaShengxiang, LiXijian. Study on prediction model of coal mine gas emission by improved bp neural network[J]. Mining Research and Development, 2019, 39(10): 138-142.
YangChao, ZhouWenzheng, LiuYuzhu. Prediction model of gas emission based on adaptive enhancement optimization[J]. Journal of Liaoning Technical University(Natural Science), 2023, 42(6): 733-739.
ZhengYongqi, ZhangLin, GaoShengqiang, et al. Photovoltaic power output prediction based on NRBO-FMD-BiGRU-Attention[J]. Science Technology and Engineering, 2025, 25(30): 12954-12963.