Study on influencing factors of gas emission in mining face based on regression shapley value decomposition method

Abstract

With the increasing of mining depth, the underground environment becomes more complex and so it is urgent to determine the influence degree of various factors on the change of gas emission in the mining face. The Shapley value decomposition method, based on regression, was used to determine the contribution of each factor to the amount of gas gushing from the recovery workings. SPSS statistical analysis software was used to analyze the data to ensure that the obtained data met the prerequisite of applying multiple linear regression analysis; through stata statistical analysis software, OLS regression method was used to model the modified data and establish the regression equation. Based on the Shapley value decomposition method, the regression equation is decomposed, and the decomposition results are verified by the verification group data. On the premise that the occurrence condition of coal seam is super thick coal seam and the quaternary stratum is mainly loess, it is concluded that the contribution rate of each factor affecting the gas emission of the mining face is as follows: coal seam thickness > advancing speed of the working face > gas content of the coal seam > the thickness of the overlying bedrock of the coal seam, and the thickness of the coal seam is the decisive factor affecting the gas emission of the mining face. By summarizing the thickness of coal seam, advancing speed of working face, gas content of coal seam and thickness of overlying bedrock of coal seam as natural factors and mining factors, it is concluded that natural factors are the decisive factors affecting the amount of gas emission in working face.

Keywords

• regression model
• shapley value decomposition
• gas gushing from mining face
• spss statistical analysis software
• stata statistical analysis software