The identification of coal and gangue and the prediction of the degree of coal metamorphism based on the EDXRD principle and the PSO-SVM model

Abstract

in characterizing the the real-time investigates coal and gangue identification and the prediction of the coal metamorphism degree based on the X-ray diffraction principle. Major coal varieties as well as gangue in the huainan were studied in this and both ADXRD and EDXRD patterns of the samples were collected. Differences in the physical phase informa tion the EDXRD patterns of and gangue are analyzed to extract the identification features of and gangue to achieve the identification of and gangue by the PSO-SVM model. Then, the technique characterizing metamorphism ing technology mainly focus on the identification of coal and gangue, and no in-depth research has been carried out on the identification of coal species. Accordingly, in order to preliminary screen coal types, this paper proposed a method to predict the coal metamorphic degree while identifying coal and gangue based on Energy Dispersive X-Ray Diffraction (EDXRD) principle with 1/3 cok ing coal, gas coal, and gangue from Huainan mine, China as the research object. Differences in the phase composition of 1/3 coking coal, gas coal, and gangue were analyzed by combining the EDXRD patterns with the Angle Dispersive X-Ray Diffraction (ADXRD) patterns. The calculation method for characterizing the metamorphism degree of coal by EDXRD patterns was investigated, and then a PSO-SVM model for the classification of coal and gangue and the prediction of coal metamorphism degree was developed. Based on the results, it is shown that by embedding the calculation method of coal metamorphism degree into the coal and gangue identification model, the PSO-SVM model can identify coal and gangue and also output the metamorphism degree of coal, which in turn achieves the purpose of preliminary screening of coal types. As such, the method provides a new way of thinking and theoretical reference for coal and gangue identification.