Thickness Distribution Prediction for Tectonically Deformed Coal with a Deep Belief Network: A Case Study

Abstract

Thickness of tectonically deformed coal (TDC) has positive correlations with the susceptible gas outbursts in coal mines. To predict the TDC thickness of the coalbed, we proposed a prediction method using seismic attributes based on the deep belief network (DBN) and dimensionality reduction. Firstly, we built a DBN prediction model using the extracted attributes from a synthetic seismic section. Next, we transformed the possibly correlated seismic attributes into principal components through principal components analysis. Then, we compared the true TDC thickness with the predicted TDC thicknesses to evaluate the prediction accuracy of different models, i.e., a DBN model, a support vector machine model, and an extreme learning machine model. Finally, we used the DBN model to predict the TDC thickness of coalbed No. 8 in an operational coal mine based on synthetic experiments. Our studies showed that the predicted distribution of TDC thickness followed the regional characteristics of TDC development well and was positively correlated with the burial depth, coalbed thickness, and tectonic development. In summary, the proposed DBN model provided a reliable method for predicting TDC thickness and reducing gas outbursts in coal mine operations.

Macromolecular Structure Controlling Micro Mechanical Properties of Vitrinite and Inertinite in Tectonically Deformed Coals—A Case Study in Fengfeng Coal Mine of Taihangshan Fault Zone (North China)

“…Based on the widely used values for the chemical shifts related to different functional groups of coal [40][41][42], fal (aliphatic carbons, 0-90 ppm), fa (aromatic carbons, 90-165 ppm), fa H (protonated aromatic carbons, 100-129 ppm), fa B (aromatic bridgehead carbons, 129-135 ppm), fa S (alkylation aromatic carbons, 135-150 ppm) and fa P (non-protonated aromatic carbon constituted by phenolic, 150-165 ppm) have been calculated. In addition, the XBP (the ratio of aromatic bridge carbon to aromatic peripheral carbon), which is an indicator of the basic structural unit (BSU) size [9], was calculated following XBP = fa B / (fa H + fa P + fa S ). In this study, fa / fal was also calculated to reveal the degree of enrichment of aromatic carbons to aliphatic carbons.…”

“…Based on the widely used values for the chemical shifts related to different functional groups of coal [40][41][42] been calculated. In addition, the X BP (the ratio of aromatic bridge carbon to aromatic peripheral carbon), which is an indicator of the basic structural unit (BSU) size [9], was calculated following…”

“…Coal basins in China have generally experienced intensive tectonic movements [2,3]; therefore, TDCs are widespread, hindering the safe and effective exploitation of coal resources and coalbed methane [4][5][6][7]. Therefore, TDC has been studied extensively [8][9][10].…”

Hyperparameter Tuning of Machine Learning Algorithms Using Response Surface Methodology: A Case Study of ANN, SVM, and DBN

“…DBN can be applied as a prediction model in various fields (i.e., prediction for reservoir landslide displacement and prediction for deformed coal) and some other applications such as image recognition, speech recognition, and clustering. [43,44].…”

Application of Improved Deep Belief Network Model in 3D Art Design

“…With the rapid development of the Internet and message technique, the Internet has provided users with more and more messages and services, but in the face of the exponential growth of message resources on the Internet, users have to spend a lot of time looking for a useful message for themselves [18]. In a probability graph, nodes represent variables, and edges represent the dependencies of variables [19].…”