Abstract. The risk of spontaneous combustion of coal is highly serious especially for underground coal mines. Coals react with oxygen immediately after producing in coal mines and they are subject to spontaneous combustion. This reaction is an exothermic reaction and coals give heat to the surrounding air. If the heat is not removed from the surrounding air, spontaneous combustion occurs and then flaming fire starts. In addition, in underground coal mine besides exothermic reactions, hazardous, explosive and flammable gases are released to the surrounding air. This situation causes interruptions in production, economical losses and environmental problems. In order to avoid these problems, coal production at the longwall should be stopped and the contact of the coal with air should be prevented. In this study, the temperatures changes throughout the longwall in the spontaneous combustion condition were measured by a temperature sensor placed inside the longwall. In addition, while the temperature recording was carried out for inner part of the longwall, the temperature of the exhaust air and concentration of some mine gases (CO, CH 4 , O 2 and CO 2 ) were also measured. Graham's ratio was calculated and used to determine of spontaneous combustion with the concentration of those mine gases. The temperatures of the longwall were increased in time and the coal production of longwall was stopped. After eliminating the air contact of coal, the temperature recording was continued. The temperature changing for the inner part of the longwall were observed continuously.

Companies consuming large amounts of coal should work with coal stocks in order to not face problems due to production delays. The industrial-scale stockpiles formed for the aforementioned reasons cause environmental problems and economic losses for the companies. This study was performed in a coal stock area of a large company in Konya, which uses large amounts of coal in its manufacturing units. The coal stockpile with 5 m width, 10 m length, 3 m height, and having 120 tons of weight was formed in the coal stock area of the company. The inner temperature data of the stockpile was recorded by 17 temperature sensors placed inside the stockpile at certain points. Additionally, the data relating to the air temperature, air humidity, atmospheric pressure, wind velocity, and wind direction that are the parameters affecting the coal stockpile were also recorded. A statistical model applicable for a spontaneous combustion event was developed during this study after applying multi-regression analyses to the data recorded in the stockpile during the spontaneous combustion event. The correlation coefficients obtained by the developed statistical model were measured approximately at a 0.95 level. Thus, the prediction of temperature variations influential in the spontaneous combustion event of the industrial-scale coal stockpiles will be possible.