Abstract: In order to study the permeability evolution of coal sample from Qinshui basin, an experimental device based on the transient pressure pulse-decay method for measuring the permeability of coal sample is developed. Permeability measurements using N2 and CO2 are conducted under common laboratory conditions which include constant pore pressure and variable confining pressures, constant effective stress and constant confining pressure and variable pore pressures. The Connell model is used to analyze the testing data. The reasons for the difference between the model prediction and the experimental value are discussed. The results show that: (1) The permeability of coal decreased with the increase of effective stress under constant confining pressure and variable pore pressures; the permeability of coal decreased with the increase of pore pressure under equivalent effective stress; as the pore pressure increased, the permeability of coal first decreased and then increased under the condition of constant confining pressure and variable pore pressures. (2) The permeability value predicted by the Connell model for constant confining pressure and variable pore pressures is greater than the experimental value. The difference between the permeability prediction and the experimental value may be due to the estimation error of the fracture compressibility and the adsorption strain coefficient. By carrying out permeability experiments and model studies in laboratory, it is of great significance to guide the experiment and simulation researches of CO2 sequestration and gas displacement in laboratory.

Key words: permeability of coal, transient pressure pulse-decay method, model prediction, fracture compressibility coefficient, adsorption strain coefficient