Abstract: In view of the problems such as large-scale dynamic disaster, high degree of destruction and difficult prevention and control in deep coal mining, the theoretical analysis method is adopted. Firstly, based on the micro-seismic data of coal and rock mass in the mining process of deep steep coal seam in Wangjiashan Coal Mine, the CMEAS algorithm is used to optimize the micro-seismic network to obtain the optimal network layout. Secondly, the multi-parameter comprehensive early warning index system of micro-seismic is constructed; according to the actual situation of the site, the risk judgment standard of monitoring data in the mining process is analyzed. Finally, under the best micro-seismic network system, based on the field micro-seismic monitoring data, the temporal and spatial evolution law of the vibration of the roof and floor strata and the front coal seam in the mining process is concluded. In addition, the numerical calculation method is used to simulate the change of displacement field and stress field of coal and rock mass in the mining process of the working face, and the activation mode of peripheral faults, thus the mechanical response law caused by the disturbance of deep mining is obtained.