Abstract: In order to solve the problems of incomplete,discontinuous,weak pertinence and inaccuracy in the prediction of coal and gas outburst risk in high-yield and high-efficiency mines in Yangquan Mining Area,and through the deep excavation of massive field data such as gas geology,gas emission and pressure monitoring in the test mines,this paper established a dynamic prediction method for two stages of coal and gas outburst with the combination of far-field and near-field by which the macroscopic gas geology quantitative grading technology and the real-time prediction technology of working face are effectively integrated,and based on the GIS platform,a computer software system for the prediction of coal and gas outburst is developed.The technology has been applied in Xinjing Mine,Xinyuan Mine,No.5 Mine,Pingshu Mine,Kaiyuan Mine and Sijiazhuang Mine in Yangquan Mining Area,and the application results show that the technology greatly improves the accuracy and advance of gas outburst prediction,for it accurately captures the actual outburst danger of working face many times,and predicts 1～2 days before cutting index method;furthermore,the technology is supported by brand-new forecasting methods,and with the help of geographic information and computer technology,and realizes rapid identification and dynamic prediction of outburst danger,and alleviates the pressure of gas prevention and control in high-yield and high-efficiency mines.