| With the increasing depth and intensity in coal mining,rock burst has become one of the major disasters faced by mining,which seriously threatens the safety of mines.it is an important basis for the prevention and control of rock burst to make the rock burst hazard analysis by analyzing geological occurrence conditions and mining technical conditions and to use the monitoring data to explore the precursory characteristics of rock burst in the face.It is of great engineering significance to ensure safe mining of mines.In this thesis,the methods and means of engineering investigation,numerical simulation and field data monitoring are used to carry out the division of rock burst dangerous area and the selection of early warning index in working face in an all-round way.The BP neural network prediction model is constructed,the whole working face is divided into small unit areas,each small unit is predicted and analyzed respectively,and the distribution of rock burst hazard grade in different areas of the working face is obtained.The FLAC numerical model is established to analyze the stress distribution and migration characteristics of different mining distances in the working face,and to explore the main dangerous area of rock burst.when the working face is under the solid coal,the middle and lower part of the working face is the main concentration area of high stress.The “triangle”of high stress is formed,and the phenomenon of stress superposition appears near the goaf of W1145,and the peak value of vertical stress reaches 41.2 MPa at 666 m;When it is below the goaf,the stress moves from the middle-lower part to the middle-upper part,forming a high stress “elliptical”area.The analysis of microseismic data is carried out,and the precursory characteristics of microearthquake occurrence of rock burst are disclosed.before rock burst occurs,The overall variation trend of daily total energy-frequency is “V” type,“N”type and “ascending”type.rock burst events occur in the upward stage of the change trend,and the high frequency area of rock burst events occurs near 100 m ahead of the working face.Before rock burst,There is an obvious low density region of energy events in the spatial distribution of microseismic events.By using trend method,critical value method and dynamic change rate method,the early warning index of rock burst in Kuangou coal mine is determined,and the comprehensive early warning method of rock burst suitable for Kuangou coal mine is established,and the field application is carried out,and the effect is good.This study has good scientific significance and reference value for promoting mine rock burst hazard analysis and early warning of similar geological conditions and mining technical conditions. |