Identification of coal and gangue based on R value method for dual-energy X-ray of Geant4 simulation

Abstract

With technical means of Geant4 simulation, the effectiveness of R value method in coal and gangue identification method of dual-energy X-ray is verified, and the characteristic factors affecting R value are explored. In the experiment, control variable method is used to study the effect of the thickness and shape of coal and gangue in R threshold. Compared with the shape of the coal sample, the thickness of the sample can directly affect the changes in R value of coal and gangue more. In addition, R threshold is continuously changing during the process of coal and gangue separation. R value of coal and gangue is positively correlated with the thickness of the sample, while the correlation with shape is difficult to be expressed due to the difficulty of mathematical descriptions of shape. The results show that the combination of dual-energy X-ray and image processing of Geant4 simulation can effectively identify coal and gangue, and the accuracy of distinguishing coal and gangue with different shapes and thicknesses reaches 95%. Finally, the current use of Geant4 simulation based on sample thickness can establish R threshold change curve for coal and gangue separation under specific geological environment for further coal and gangue separation with R value method based on dual-energy.

Keywords: Coal; Dual-energy X-ray; Gangue; Geant4 simulation; R value.