[Objective] Spontaneous coal combustion is one of the major natural disasters in coal mining; thus, accurate prediction of the risk of spontaneous coal combustion is crucial to prevent and control coal fire disasters. However, the complexity of the physicochemical process of spontaneous coal combustion and its various influencing factors poses a challenge to the risk prediction of spontaneous coal combustion. Strengthening research on spontaneous coal combustion hazard prediction technology using deep learning is crucial for improving the intelligent control level of coal mine safety production.[Methods] In this study, CO volume fraction was chosen as the index for spontaneous coal combustion evaluation. A dataset was constructed, and the field observation data were visualized. Next, the dataset was tested for the distribution of eigenvariables, normalized for the distribution of eigenvariables, and normalized for the dataset using kernel density estimation, logarithmic transformation, and maximum-minimum normalization. Finally, three algorithms, namely recurrent neural network (RNN), long short-term memory (LSTM) network, and gated recurrent unit (GRU), were applied to the data mining of spontaneous coal combustion feature information, and a dynamic sequence prediction model of spontaneous coal combustion CO volume fraction was established. During the model construction process, the full connectivity layer and Dropout class were added to prevent overfitting, and the mean square error and three model performance test indicators were introduced to analyze and optimize the model parameters and test the model performance.[Results] The results were presented as follows:(1) The CO volume fraction sequence dataset was established based on the field data of the Dafosi Coal Mine, the model generalization capability was enhanced, and the training time of the model was shortened by preprocessing the dataset. (2) The RNN, LSTM, and GRU models achieved the dynamic prediction of CO with an error of less than 1 %. (3) The optimal parameters of the three models were determined from the mean absolute error (MAE), the root mean square error (RMSE), and R2 of the training and validation sets. A comparative study using the model performance evaluation metrics revealed that the LSTM model had the highest prediction accuracy under the same sequence data, followed by the RNN and GRU models.[Conclusions] Using 285 sets of field data, the spontaneous coal combustion CO volume fraction sequence prediction models based on the RNN, LSTM, and GRU algorithms were established. The experimental values of the CO volume fraction were highly consistent with the predicted values, and the prediction error was less than 1 %. The model can predict the change in the CO volume fraction in future moments using the dataset. The results reveal that the dynamic time series prediction of CO volume fraction from spontaneous coal combustion using sequence models is possible compared with conventional static models. Moreover, the process of constructing the three models and optimizing the parameters can be employed as a basic study for developing sequence prediction models for other indicator gases.
Key words
spontaneous coal combustion /
CO volume fraction prediction /
long short-term memory (LSTM) network /
deep learning
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
References
[1] VAN DIJK P, ZHANG J Z, JUN W, et al. Assessment of the contribution of in-situ combustion of coal to greenhouse gas emission; Based on a comparison of Chinese mining information to previous remote sensing estimates[J]. International Journal of Coal Geology, 2011, 86(1):108-119.
[2] LIANG Y C, LIANG H D, ZHU S Q. Mercury emission from coal seam fire at Wuda, Inner Mongolia, China[J]. Atmospheric Environment, 2014, 83:176-184.
[3] LIU S Q, WANG C H, ZHANG S J, et al. Formation and distribution of polycyclic aromatic hydrocarbons (PAHs) derived from coal seam combustion:A case study of the Ulanqab lignite from Inner Mongolia, northern China[J]. International Journal of Coal Geology, 2012, 90-91:126-134.
[4] O'KEEFE J M K, HENKE K R, HOWER J C, et al. CO2, CO, and Hg emissions from the Truman Shepherd and Ruth Mullins coal fires, eastern Kentucky, USA[J]. Science of the Total Environment, 2010, 408(7):1628-1633.
[5] ENGLE M A, RADKE L F, HEFFERN E L, et al. Gas emissions, minerals, and tars associated with three coal fires, Powder River Basin, USA[J]. Science of the Total Environment, 2012, 420:146-159.
[6] KUENZER C, STRACHER G B. Geomorphology of coal seam fires[J]. Geomorphology, 2012, 138(1):209-222.
[7] 肖旸,王振平,马砺,等.煤自燃指标气体与特征温度的对应关系研究[J].煤炭科学技术, 2008, 36(6):47-51. XIAO Y, WANG Z P, MA L, et al. Research on correspondence relationship between coal spontaneous combustion index gas and feature temperature[J]. Coal Science and Technology, 2008, 36(6):47-51.(in Chinese)
[8] 邓军,白祖锦,肖旸,等.煤自燃指标体系试验研究[J].安全与环境学报, 2018, 18(5):1756-1761. DENG J, BAI Z J, XIAO Y, et al. Experimental investigation and examination for the indexical system of the coal spontaneous combustion[J]. Journal of Safety and Environment, 2018, 18(5):1756-1761.(in Chinese)
[9] CHEN J C, LI L, JIANG D Y, et al. Experimental study on the spatial and temporal variations of temperature and indicator gases during coal spontaneous combustion[J]. Energy Exploration&Exploitation, 2021, 39(1):354-366.
[10] 邓军,肖旸,陈晓坤,等.矿井火灾多源信息融合预警方法的研究[J].采矿与安全工程学报, 2011, 28(4):638-643. DENG J, XIAO Y, CHEN X K, et al. Study on early warning method of multi-source information fusion for coal mine fire[J]. Journal of Mining and Safety Engineering, 2011, 28(4):638-643.(in Chinese)
[11] ZHAO J Q, YANG D G, WU J X, et al. Prediction of temperature and CO concentration fields based on BPNN in low-temperature coal oxidation[J]. Thermochimica Acta, 2021, 695:178820.
[12] MENG Q, WANG H Q, WANG Y S, et al. SVM based prediction of spontaneous combustion in coal seam[C]//Proceedings of 2008 International Symposium on Computational Intelligence and Design. Wuhan:IEEE, 2008:254-257.
[13] GUO Q, REN W X, LU W. A method for predicting coal temperature using CO with GA-SVR model for early warning of the spontaneous combustion of coal[J]. Combustion Science and Technology, 2022, 194(3):523-538.
[14] SAFDARNEJAD S M, TUTTLE J F, POWELL K M. Dynamic modeling and optimization of a coal-fired utility boiler to forecast and minimize NOx and CO emissions simultaneously[J]. Computers&Chemical Engineering, 2019, 124:62-79.
[15] ZHAO H Q, LU L, HE Z Y, et al. Adaptive recursive algorithm with logarithmic transformation for nonlinear system identification in α-stable noise[J]. Digital Signal Processing, 2015, 46:120-132.
[16] FENG Q H, ZHANG J Y, ZHANG X M, et al. Proximate analysis based prediction of gross calorific value of coals:A comparison of support vector machine, alternating conditional expectation and artificial neural network[J]. Fuel Processing Technology, 2015, 129:120-129.
[17] LAUBSCHER R. Time-series forecasting of coal-fired power plant reheater metal temperatures using encoder-decoder recurrent neural networks[J]. Energy, 2019, 189:116187.
[18] GOODFELLOW I, BENGIO Y, COURVILLE A. Deep learning[M]. Cambridge, USA:MIT Press, 2016.
[19] SCHMIDHUBER J. Deep learning in neural networks:An overview[J]. Neural Networks, 2015, 61:85-117.
[20] GÉRON A. Hands-on machine learning with scikit-learn and tensorflow:Concepts, tools, and techniques for building intelligent systems[M]. Beijing:O'Reilly Media, 2017.
[21] GREFF K, SRIVASTAVA R K, KOUTNÍK J, et al. LSTM:A search space odyssey[J]. IEEE Transactions on Neural Networks and Learning Systems, 2017, 28(10):2222-2232.
PDF(13984 KB)
