Study On Precursor Information Identification And Early Warning Model Of Coal Mine Rock Burst

Rockburst is one of the coal mine power disasters and is extremely destructive.At present,China has reached 253 rockburst mines,and as coal resources are tapped into the deep,the situation of rockburst mine pressure will become more severe,which seriously affects the safe mining of coal resources in China.Monitoring and early warning of rock burst pressure is of great significance to disaster prevention and decision-making in coal mines,and is an important field in the research of rock burst pressure.In this paper,we use theoretical analysis,laboratory study,numerical calculation,engineering practice and other methods to study the characteristics of precursor information of coal and rock instability and damage,establish a comprehensive early warning model of rock burst based on microseismic method,and test the early warning effect of the early warning model.This paper has formed the following main results:Theoretical analysis of the mechanical basis of early warning of rockburst,through the monitoring and early warning of energy can be achieved early warning of coal and rock damage.The principle of monitoring and early warning of strong mine-shock caused by coal mining excavation and the use of mine-shock time sequence,space,and intensity indicators are analyzed,and the physical significance and calculation methods of 11 mine shock precursor indicators are given.Through laboratory tests,the spatiotemporal strength characteristics of coal sample damage rupture acoustic emission events and the coupling relationship between coal sample damage rupture and precursor indicators were studied.The results show that: the sample is in the elastoplastic stage is the main energy release stage;H-1 ~ H-6 sample count and energy accumulation curve can be seen as "stepped" or "slow stepped",this period can be regarded as a precursor of coal sample instability and destruction;EEM and b,A(b)value,mine shock activity S,total fault area A(t)and Z-map index have obvious precursor characteristics before coal sample destruction,and can be used as early warning indicators for coal sample instability and destruction.Based on the mine shock data of the 401105 working face of Hujiahe Coal Mine,the spatial and temporal strength analysis of the mine shock was carried out to obtain the spatial distribution characteristics of the mine shock.The high-energy mine shock events are separated in the occurrence period and there is the rule of "shockless" phenomenon.The numerical calculations have obtained six early warning indicators of A(b),A(t),mine shock activity S value,EEM,b value,and Z-Map.By comparing the results of acoustic emission and on-site mine shock,it is concluded that the temporal and spatial distribution,energy changes,and precursor indicator changes of events before coal body instability and destruction at two scales show similar trends.Based on the precursory indicators of mine shock,the reliability theory is combined with the index standardization method to normalize the dimensions of the indicators,and the membership functions of each early warning level of the precursory indicators of mine mine are established using Gaussian membership functions.The F-score value in the confusion matrix is used to evaluate the early warning effect of each mine shock precursor indicator,and the weight of each precursor indicator is calculated.Based on the fuzzy mathematics theory,a method for quantitatively evaluating the risk of shock is established,and the principle of maximum membership(MMDP)and variable fuzzy pattern recognition(VFPR)are used to ensure the reliability of the comprehensive early warning results.The reliability and practicability of the comprehensive early warning model of rockburst pressure were verified by the observation results of the mine shock in the 401105 working face of Hujiahe Coal Mine.The early warning accuracy of each indicator in the comprehensive early warning model is: EEM> b value> A(b)value> mine shock activity S> total fault area A(t)> Z-map value,and the early warning model is compared The early warning effect is better than the single indicator early warning.The new mine shock data was used to test the early warning effect of the early warning model on strong mine shock events,and the accuracy rate of the comprehensive early warning model for strong mine shock events was 80%.This paper has 64 pictures,10 tables,and 96 references.