The blade erosion of a tidal current turbine can lead to significant energy losses and affect stall behavior. To maintain good performance and prevent turbine malfunction, it is important to determine the location and rate of blade erosion caused by particle impact. In this study, a computational fluid dynamics and discrete phase model (CFD–DPM) method is employed to study the erosion characteristics in blades subjected to multiphase flow. The fluid–particle interactions and the influence of turbulence on the particle trajectories are considered in the CFD–DPM model. The maximum erosion location and the average erosion rate are investigated under different particle diameters, particle concentrations, particle shape factors, and airfoil parameters. The fluid velocity, particle velocity, and particle trajectory are further analyzed to reveal the erosion mechanism under different influencing factors. The results show that while both the maximum erosion location and the average erosion rate depend upon particle independence (the greater the degree to which particles deviate from the fluid streamline, the better the particle independence), the latter is also related to the drag force exerted by the particle. For a tidal current turbine, erosion occurs first at the tip and leading edge of the blade, and the most severe erosion area is the blade tip. The erosion laws obtained in this work can provide guidance for erosion prediction, tidal current turbine field site selection, and blade optimization.