Experimental Study On Deformation And Permeability Of Gas-bearing Coal(Rock) Under Dynamic Load

With the continuous increase of coal production,the mining depth is deepening,the geological conditions are becoming increasingly complex,and the coal mine safety situation is becoming increasingly severe.The dynamic disasters such as mine earthquake and their secondary disasters are becoming more and more severe,which pose a great threat to personal and property safety.Due to the limitation of test instruments,sealing means and testing methods,it is impossible to quantitatively study the mechanical deformation characteristics and seepage characteristics of coal samples under dynamic load after adsorption.Therefore,the current research mainly focuses on the mechanical deformation and seepage characteristics of coal and rock under static load,and there is relatively little research on the prominent disasters caused by dynamic disaster factors such as mine earthquake.Therefore,it is of great significance to study the deformation and permeability characteristics of coal strata simulated by mine earthquake in laboratory for the early warning work of coal mine gas outburst and other disasters.In this paper,the gas desorption and permeability of coal rock under the action of mine earthquake are simulated by an improved GDS dynamic triaxial test system,the influence of coal mine earthquake on coal seam deformation and permeability is discussed.The dynamic characteristics and permeability of gas bearing coal under cyclic dynamic load are studied,CSR and load frequency on coal sample deformation and permeability,the mechanical properties of gas carrying coal under different conditions and the general rule of leakage are revealed,and establish the permeability prediction model of coal sample based on BP neural network optimization algorithm.The main research contents and achievements are as follows:(1)Aiming at the problem that the current dynamic load test technology cannot meet the test requirements,improve the design and processing of the GDS dynamic triaxial test system and add a gas adsorption system;Based on the analysis of gas adsorption mechanism,the feasibility of indoor simulation test with CO2 instead of gas is discussed,and the indoor simulation method of coal rock gas release under the action of coal earthquake is put forward.(2)The mechanical properties of gas samples at different pressures,CSR and loading frequencies are discussed.Under different deposition pressures,the peak deformation of coal samples is negatively related to the deposition pressure,and the axial deformation capacity decreases with the increase of deposition pressure.Under different CSR conditions,the deformation value is positively correlated with the change of CSR,and CSR has the greatest influence on the coal sample.Under different load frequency conditions,the strain is negatively correlated with the load frequency,and the transverse deformation capacity decreases with the increase of load frequency.(3)By analyzing the leakage characteristics of samples under different deposition pressures,CSR and loading frequencies,the changing laws of permeability,permeability and permeability are given.The permeability of the sample decreases with the increase of pressure,and the gas seepage rate increases with the increase of CSR,which has a negative linear correlation with the load frequency.The seepage rate of gas-bearing coal has a negative linear correlation with confining pressure,a negative exponential function with load frequency and a positive exponential function with CSR.Under different confining pressure conditions,permeability decreases with the increase of confining pressure,which is positively correlated with CSR and negatively correlated with frequency.(4)According to the permeability data under different pressure,CSR and cyclic load,the BP neural network permeability prediction model is established,and the BPGA neural network training is optimized using genetic algorithm,parameter processing and superparameter selection process are described.The accuracy of the model is described by evaluating parameters,and the predicted value is close to the measured value,which indicates that the model can better predict the permeability of coal samples.