Investigation on the Quantitative Evaluation Method of Coal Combustion Intensity in O2-Co2-N2 Atmospheres Based on Dynamic Artificial Neural Network

The spread of coal combustion leads to a large-scale, continuous and violent combustion of coal seams, called coalfield fires, resulting in a colossal waste of energy. The multi-gas mixing atmospheres underground affect the coal combustion intensity in different degrees, thus affect the spread of a coalfield fire. In this paper, simultaneous thermal analysis experiments were conducted to investigate the combustion behaviors of high volatile bituminous coal and anthracite in O2 -CO2 -N2 atmospheres with eight groups of mixing ratios. A new index for quantitative characterization of coal combustion intensity in multi-gas mixing atmospheres was proposed. Furthermore, a method to predict and judge the coal combustion intensity by artificial neural network (ANN) was systematically proposed. Results showed that the combustion rate decreased obviously with the CO2 ratio increasing, and the O2 ratio had a greater impact on coal combustion than the CO2 ratio. The gas mixing ratios mainly affect the burnout temperature of coal combustion. In the 15%O2 -5%CO2 -80%N2 atmosphere, the combustion could be inhibited for high volatile bituminous coal, while be promoted for anthracite. Based on the proposed coal combustion intensity index and the selected optimal ANN model, the coal combustion intensity could be well predicted