Abstract: Most of the existing studies on CO2 replacement for CH4 are focused on microscopic gas injection replacement mechanism and macroscopic replacement efficiency influencing factors, and most of the studies are theoretical analysis at the model level or multiple factor analysis at the simulation test level. There are few physical simulation experiments and quantitative analysis. In order to solve the above problems, by using CO2 replacement for CH4 experiment system, this paper studies the seepage diffusion evolution law and time-varying characteristics of CO2 replacement for CH4 in coal during the replacement process at different injection pressures. This paper analyzes the change law of CO2 and CH4 concentrations at the gas outlet, the accumulated CH4 replacement volume and the replacement ratio during the whole process. Physical simulation experiments are used to investigate the influence of gas injection pressure on the replacement efficiency and make the quantitative analysis. The experiments results show that: ① The CO2 and CH4 gas concentration change trends at different injection pressures are basically the same, which can be divided into three stages, including the original equilibrium stage, the dynamic equilibrium stage and the new equilibrium stage. As the gas injection pressure increases, the time of CO2 and CH4 gas breaking the original equilibrium stage is gradually shortened, the time of the dynamic equilibrium stage increases, and the time of the new equilibrium stage is about the same. ② At different injection pressures, the accumulated CH4 replacement volume increases with the increase of gas injection time, and the increase rate is fast at first and then slow, and becoming a fixed value at last. The gas injection pressure increases from 0.6 MPa to 1.4 MPa, the accumulated CH4 replacement volume increases, the replacement ratio decreases from 4.99 to 4.10, and the replacement efficiency increases. When the gas injection pressure is 1.4 MPa, the replacement ratio is the smallest and the replacement efficiency is the best. This conclusion can provide a reference for the theoretical study of CO2-ECBM related technologies and the selection of related process parameters for technical implementation in low permeability coal seams.