Abstract: According to the problems existing in the prediction of gas emission in small fault structure area, this paper determines the gas parameters before and after the mining in the small fault tectonic area of Wu-yang mine, Chang-cun mine, Li-cun mine and Yu-wu mine in Lu'an mining area, then, the law of gas emission in the small fault tectonic area is analyzed and the main factor index system of the small fault tectonic area which affects the gas emission is established. Using the BP neural network analysis method, the relationship between geological structure of small fault tectonic and gas emission is studied, furthermore, a prediction model for gas emission in small fault structure area is established based on BP neural network. The application in Lu'an mining area shows that, the error between the method and the measured data is less than 5%, which can effectively predict the gas emission at different locations in the small fault structure area.