Experimental study on graded early warning model and critical discrimination criteria for coal spontaneous combustion in Chahasu Mining Area

To establish a highly applicable graded early warning model for coal spontaneous combustion, a temperature-programmed experimental method was used to investigate the effect of mine ventilation rate on gaseous compound release during coal low-temperature oxidation. The characteristic temperatures of coal spontaneous combustion were determined via growth rate analysis, and a graded early warning model with corresponding critical discrimination criteria was proposed. The results show that ventilation rate significantly affects the release patterns of gaseous compounds during coal spontaneous combustion, and a single gas indicator cannot meet the requirements of coal spontaneous combustion early warning. To reduce the impact of ventilation rate on early warning performance, the coefficient of variation method combined with weighted grey relational analysis was adopted to evaluate and optimize the gas indicators. CO2/CO was determined as the primary indicator, while O2/(CO+CO2) and CO2/O2 were selected as alternative indicators. The Logistic fitting curves of the indicators were used to infer the hazard development trend, and coal spontaneous combustion early warning was divided into three levels: green, yellow and red. A graded early warning indicator system for coal spontaneous combustion was established, and the proposed critical discrimination criteria can quickly determine the risk level of hazard areas based on underground gas monitoring data, which is of great significance for precise early warning and effective prevention of coal spontaneous combustion in the field.