The application of BP neural networks in the Study of influence factor of CBM production capacity

Abstract

The exploitation of coal seam gas is influenced by various geological factors. In order to study the weights of these factors, this paper establishes a BP neural network prediction model by taking No.3 coal seam in the representative area of southern Qinshui basin as an example. It finds out that the potential coal-seam gas capacity of southern area of Qinshui basin is influenced by some primary factors as follows: the depth of the coal seam, the thickness of the coal seam, the content of the gas, the degree of permeability, the reservoir pressure, the lithology of root floor, geological structures, ash content and so on. Among these factors, the degree of permeability, the content of gas, and the geological structures weight higher than the other factors, and the depth of coal seam is the least influence factor. The research findings are consistent with the actual mining emissions of CBM, which explains that BP neural networks prediction model has a strong nonlinear approximation ability, which can accurately evaluate the nonlinear relation between CBM exploration potential and main control factors.