Gas disasters are the main threat to coal production safety, to achieve stable and reliable security situation prediction is very important. Effective analysis of coal mine production safety status is of great significance to coal mine safety production. This paper uses the gas monitoring data of the working face of Shendong mining area as the sample, by building a prediction model and using a mathematical model method to determine the safety level of the coal mine area. The prediction of the gas emission law of coal mining face is realized to predict the gas data trend, use the data to measure and use the root mean square error and average absolute error indicators to measure the accuracy of the gas emission prediction results, verify that the prediction accuracy rate can reach 98.21%. The establishment of a coal mine gas disaster prediction model is of great significance for the avoidance of coal mine risks. Predicting possible coal mine gas disasters in advance can better ensure the safety of coal mine operations and reduce the property and life and property losses of mine and underground personnel minimum.