In order to forecast network security situation effectively, puts forward a forecast method which combines GM(1,1) and GM(1, N) model on the basis of situation factors and grey theory. First filters situation factors, then applies model GM(1,1) to forecast variation of the situation factors, and obtains N functions of situation factors variation, finally uses the functions and grey model GM(1, N) to forecast network security situation. The grey model GM(1,1), neural network model and the proposed method are used to forecasts network security situation, and the experimental results prove that the proposed method forecasts more accurately the network security situation.