Abstract: Because the complicated non-linear relation between the coal-gas-outburst and its affecting factors,it is difficult to established model with traditional mathematical method.The authors set up the forecasting system with fuzzy-neural network theory,by means of fuzzy mathematics to express and deal with the imprecise data and fuzzy information,and utilized self-adaptive neural networks system to solve the problems.The results of actual application indicate that the model is reliable and precise and has large practical value,which enriches the theory of forecasting coal-gas-outburst.