Classification of Coal Seams with Respect to their Spontaneous Heatingg Susceptibility using K-means Clustering

Abstract

The current paper presents the k-means clustering approach for classification of coal seams with respect to their spontaneous heating susceptibility. To apply this technique, a number of coal samples of varying ranks have been collected from different coalfields of the country. These samples include some of the coal seams, which are well known for their high and low susceptibility as observed in the mines. The intrinsic properties of the coal seams have been determined by proximate, ultimate and petrographic analyses. The spontaneous heating proneness of the samples has been studied using different experiments in the laboratory, viz. crossing point temperature, differential thermal analysis, critical air blast test and differential scanning calorimetry. Correlation studies between the intrinsic properties and susceptibility indices have been carried out to identify the parameters for clustering purposes. The identified parameters have been used as inputs in the k-means clustering algorithm and coal seams have been classified into four different categories. This classification approach will help the planners and mining engineers in the field to adopt appropriate measures for preventing the occurrence of mine fires