Experimental assessment of multi-parameter index gas correlation and prediction system for coal spontaneous combustion

Investigating the variation pattern of indicator gases produced during the spontaneous combustion and oxidation of coal is of crucial value in achieving early warning of coal mine fires and ensuring coal mine safety. In this work, we first carried out a programmed heating experiment on coal and mainly analysed the trend of oxidation products, alkane gas, alkene gas, and the ratio of each indicator gas with temperature. Subsequently, the relational degree between the 13 indicator gases or indicator gas ratios and coal temperature was calculated by combining grey correlation analysis (GRA). The ranking of each indicator to match the temperature trend was determined by the correlation magnitude, and the top three ranking indicators suitable for predicting coal spontaneous combustion (CSC) at five different temperature stages were reasonably selected. Following this, a mathematical function model between the data values of the coal spontaneous combustion indicators and the temperature was developed and the temperature ranges that could be predicted by indicators were calculated. Finally, the absolute errors of the upper and lower forecast limits of the indicators were discussed in detail. It was found that the upper and lower limit values for the first indicator were smaller than those for the second and third indicators, which also which also confirmed the accuracy of the first indicator. The top three predictors at different temperature stages were compared with historical data and the first indicator gases were consistent between 30 °C∼50 °C and between 50 °C∼90 °C, being CO and C2H6 respectively. Physically, CSC is a very complex oxidation process and the predicted indicator gases for the same coal sample may vary. In actual coal mine production, there is still a necessity to further develop and improve the multi-gas prediction index system reflecting the degree of coal spontaneous combustion in order to reduce the losses caused by mine fires.