In this paper, on the basis of analyzing the changing laws of coal and gas outburst premonition electromagnetic emission (EME), by means of the theories of system identification, dynamic testing data processing and mathematical statistics, a coal or rock rheological fracture EME anomaly judging model is established and an EME anomaly judging criterion is put forward. The model has been applied on forecasting the danger of working face coal and gas outburst. The results show that the model has a higher forecast accuracy, can avoid some disadvantages of the critical indexes forecast and realize a continuous dynamic trend forecast of the danger of coal and gas outburst in the working face.