Statistical analysis to establish an ignition scenario based on extrinsic and intrinsic variables of coal seams that affect spontaneous combustion

Abstract

Abstract Spontaneous combustion of coal seams in open pit mines is a problem that occurs in coal mines around the world. Theories have been developed that express several factors which generate self-heating in the coal and favor combustion. The synergy of these factors contributes to the increase in the temperature of oxidation reactions, facilitating the culmination of the activation energy of the system, in which the ignition develops and the reaction is spontaneous. Currently, the incidence of each variable’s susceptibility to coal combustion is unknown, so in this work a methodology was developed to determine statistically which of the variables has the greatest contribution in the phenomenon. Descriptive statistics, atypical values detection, principal component determination, cluster analysis, and logistic regression were used to determine how many indicator variables are necessary to describe susceptibility to coal combustion. For the analysis, 21,000 data from different seams of two open-pit coal mines in endwalls and abandoned pits were processed with consideration given to extrinsic and intrinsic factors. Seam temperature, atmospheric pressure, wind speed, oxygen, methane, height, ash, volatile matter, heat power, vitrinite, liptinite and the Hardgrove Index are the most influential variables in the phenomenon.