Fusion Dual-Factor Model for Quantitative Evaluation of Coal Self-Ignition Tendency

Coal spontaneous combustion is common in mining and utilization processes. Mine fires induced thereby not only result in massive economic losses but also severely threaten the ecological environment and operational safety. However, single indicators like the traditional chromatographic oxygen absorption method fail to fully encompass coal's entire oxidation process, often leading to evaluation deviations. To address this, this study proposes a microstructural characteristic-driven dual-factor coupling model (MD-DFC) for predicting coal spontaneous combustion tendency.First, by combining chromatographic oxygen absorption and crossing point temperature experiments, oxidation characteristics of coal in the low-temperature adsorption and high-temperature exothermic stages are evaluated separately. The entropy weight method is used to construct a Dual-Factor Coupling Index (DFC), overcoming the limitations of single-index approaches. Then, microscopic features such as key functional group contents are identified via Fourier transform infrared spectroscopy, and Pearson correlation analysis reveals their fundamental regulatory effects on spontaneous combustion tendency.On this basis, functional group distribution characteristics are incorporated into DFC to build the MD-DFC prediction model, enabling accurate evaluation of coal spontaneous combustion tendency at the molecular scale. Comparative experiments on typical coal samples demonstrate that MD-DFC's predictions are highly consistent with natural ignition periods, with higher sensitivity and accuracy, especially for medium-metamorphic coals. Compared to traditional methods, MD-DFC has shorter assessment time and higher accuracy, showing strong engineering applicability and promotion potential.