Abstract:Artificial neural network is of extremely strong ability in nonlinear approach, it is a available way theoretically to imitate the complicated and non-linear relations between the incident of the coal and gas outburst and its influence factors. An optimized BP network general program used for the prediction of coal and gas outburst has been developed by use of object-oriented programming and C++. In this paper, the structuring technology and design method of the program are described in detail. The practical application has proved that the of the above- mentioned trains of thought is feasible.