The proton exchange membrane fuel cell (PEMFC) has many advantages, such as low operating temperature, fast reaction, and high energy density, etc. It is a very potential energy source. During battery operation, PEMFC will produce a large amount of water in the catalyst layer where the reaction occurs. The catalyst layer and gas diffusion layer (GDL) inside the battery are made of porous media. The water produced during the reaction will be discharged through these porous media structures, but the narrow gaps of the porous media structures will complicate the process. For example, too much water will flood the internal structure of the battery, making liquid water unable to drain smoothly. The above process may cause the reaction rate of the battery to decrease or even stop. Studying and understanding the water transportation process inside PEMFC is of great help in solving the above problems. Lattice Boltzmann method (LBM) is a promising numerical method for simulating the flows in porous media. Compared with the traditional CFD method, the lattice Boltzmann method (LBM) can deal with the pore structure of fine-scale better. In this paper, the lattice Boltzmann method (LBM) is employed to simulate the flow in the PEMFC. In order to reconstruct the porous media structure in PEMFC, the numerical reconstruction technique is used to generate porous media structures with different porosity. By simulating the flow of water in porous media, the transport laws of water in different porosity rates can be obtained. It provides a reference for improving the operating efficiency of proton exchange membrane fuel cells.