Abstract: In order to explore the causes and prediction techniquesof CO emission in igneous rock intrusion area,and the effect of high temperature on coal structure and natural oxidation capacityduring igneous rock intrusionis studied by simulating the combination of thermal environment and coal samples in actual igneous rock affected area. The coal samples with different temperature pretreatments were used to simulate the magmatic intrusion of thermal metamorphism, and mercury intrusion experiments and coal CO adsorption tests were carried out to investigate the effects of thermal environment on coal metamorphism, CO adsorption and pore structure. The coal samples from three working face with different igneous rock intrusion distances were selected for comparative experiments. Temperature rise programming, infrared spectroscopy, thermogravimetry and other tests were carried out to investigate the CO release law of coal and the cause of temperature rise oxidation.The occurrence of CO in the igneous zone affected by Linnancang mine and its natural ignition prediction indexwere determined. The test results show that at the experimental temperature (300 ℃), the thermal action is the main reason for the increase of coal rank. The increase of temperature can increase the number of micropores and macropores in coal, and increase the surface area and pore volume of coal, and the pore connectivity is also enhanced;Coal sample closed to the igneous rock intrusion is easier to release a large amount of CO at a lower temperature, and it is more likely to enter the violent oxidation stage.Therefore, it can be judged that the high temperature environment of the igneous rock intrusion increases the capacity of the coal body and enhances the pore connectivity.The way of changing the structure of coal body improves the ability of CO to be attached to coal and the ability of oxygen to contact with coal in natural oxidation.It is proposed to use φ(CO2)/φ(CO) as a natural ignition prediction index for Linnancang mine.This indicator solves the problem of forecasting natural fire on site and has a high application value for coal spontaneous combustion control of similar coal mines.