The permeability of high water-cut reservoirs will change during long-term water flooding. The traditional parameters characterizing water flooding do not consider the influence of critical flow velocity when describing the time variation law of reservoir permeability. Therefore， the concept of effective cumulative water flux was put forward， and the critical flow velocity was used as the velocity limit of particle migration to correct the influence of water flux on reservoir permeability to a certain extent. The ridge evolution model of reservoir permeability changing with effective cumulative water flux was obtained by experiments and coupled with the three-dimensional two-phase oil-water flow equation. As a result， a time variation reservoir permeability flow model based on effective cumulative water flux was established. The results show that the permeability of the Berea core changes greatly even at a low displacement rate under long-term water flooding， and the critical flow velocity under high multiple water flooding （0.006 cm/s） is lower than that under normal conditions （0.015 cm/s）. By considering the effective cumulative water flux of critical flow velocity to characterize the time variation characteristics of reservoir permeability， the numerical simulation results of complex fluvial reservoir models are more consistent with the time variation characteristics of reservoir permeability caused by microparticle migration. Therefore， the effective cumulative water flux can better describe the time variation law of reservoir permeability under the influence of critical flow velocity.