Abstract: In order to solve the problem of the lack of comparative analysis of different coal low-temperature oxidation kinetic parameters test methods in existing studies, three test methods, Coats-Redfern method, q/m method and Starink iso-conversional method, are used to calculate coal low-temperature oxidation kinetic parameters respectively. The accuracy of the three methods is compared with coal oxidation self-heating reaction time and critical accumulative thickness of spontaneous combustion as comparative parameters. The adiabatic oxidation experiment, the multiple heating rate experiment and the constant temperature experiment of the coal sample under pure oxygen and poor oxygen conditions are carried out through an adiabatic oxidation device and a simultaneous thermal analyzer. Based on the results of the thermal analysis experiments, three methods are used to calculate coal low-temperature oxidation kinetic parameters. The coal oxidation self-heating reaction time and the critical accumulative thickness of spontaneous combustion are calculated based on the obtained kinetic parameters. The calculated results are compared with the measured coal oxidation self-heating reaction time and the actual thickness of residual coal in goaf to evaluate the accuracy of the three methods. The experimental results show the results as follows. ① The coal oxidation self-heating temperature change rate measured by the adiabatic oxidation experiment increases gradually with the time. The temperature calculated by the Coats-Redfern method hardly changes from 0-8 h (0-10 h for the q/m method), and the temperature increases rapidly after this time period. The temperature change trend obtained by the Starink iso-conversional method is similar to the measured result of adiabatic oxidation experiment. ② The actual thickness of residual coal in goaf is greater than the critical accumulative thickness calculated by the Coats-Redfern method. The critical accumulative thickness obtained by the q/m method exceeds the thickness of the coal seam where the working face is located, which is obviously inconsistent with the actual situation. The calculation results obtained by the Starink iso-conversional method are closer to the actual situation. ③ The prediction results obtained by the Starink iso-conversional method are closer to the measured results, indicating that the low-temperature coal oxidation kinetic parameters obtained by the Starink iso-conversional method are more accurate than the other two test methods.