2-D hybrid lattice-Boltzmann Lagrangian simulations are conducted to predict the experimental results for nanoparticle filtrations in mesh screens. Comparisons between simulations and experimental data are made for pressure drops across the screens and screen-collection efficiencies. The penetration of 10–300 nm particles through a screen is evaluated for three screen opening sizes (60, 100 and 180 µm) at different flow rates (2.5, 4 and 6 LPM). The computational results are obtained in a 2-D cross-sectional domain that represents a semi-infinite array of clean fibers. In order to simplify the 3-D configuration where two fibers interweave, three 2-D models that take into account the physical parameters of mesh screens are investigated: opening-size-, open-area- and volume-fraction-based modeling. Regarding the prediction accuracy, the results identify the most suitable mesh parameters for being used in a 2-D model to predict particle filtration. In addition, the spatial distributions of particle deposition on a single fiber are in accordance with structural growth during fiber loading in the regimes of diffusion and interception.