Abstract: Focusing on a major scientific problem to be solved,i. e. ,“research on the risk identification and monitoring and early warning technology of typical coal mine dynamic disasters”,this paper investigates the typical coal mine dynamic disasters such as coal and gas outburst and rock burst. In view of the current situation of unclear mechanism of typical coal mine dynamic disasters,unclear risk identification and warning technology of monitoring and early warning,etc. The research covers ① the development mechanism of rock burst and risk identification and monitoring and early warning,② the coal and gas outburst disaster mechanism and monitoring and early warning,③ the coal mine typical dynamic disaster signal acquisition and transmission and intelligent analysis,and ④ the coal mine typical dy- namic disaster monitoring and early warning system platform. A large-scale,true three-dimensional,fully closed and automatic experimental device for the physical simulation of coal and gas outburst is developed. A new type of sensing and fusion transmission sensor device for dynamic disaster precursor information,including fiber Bragg grating micro- seismic sensor,tri-axial stress sensor,and distributed multi-point laser methane detection,is developed. The aggregation theory and method of multi-dimensional and massive dynamic information of underground sensor data are established. The prediction method of typical dynamic disasters and the multi granularity knowledge mining method based on drift characteristics are constructed. A model of judgment,recognition and warning of major coal mine disasters based on big data analysis and data mining is established. Through the field applications,it is shown that the acquistion sensor can realize the comprehensive acquisition of man-machine ring parameters,and has the advantages of self- diagnosis on fault,short response time and long calibration cycle. The non-fault rate of the monitoring and warning sys- tem has reached 99% in a stable operation,the anti-interference level is no less than level 3,and the system's monitoring and warning accuracy is more than 90% . The system has realized the online monitoring,intelligent judgment and real-time warning of typical power hazards in coal mines.