Knowledge of porous media structure is an essential part of the hydrodynamic investigation of fluid flow in porous media. To study soil behavior (as a granular porous media) and water and contaminant movement in the vadose zone, appropriate estimation of soil water retention curve (SWRC) and soil hydraulic conductivity curve (SHCC) has a pivotal role and is one of the most challenging topics for researchers and engineers in soil and water science. The SWCR can be approximated using an accurate particle size distribution (PSD) function. In this study by applying the random close packing (RCP) method as an encouraging method for predicting and studying particle configuration, an optimal particle size distribution is developed for coarse-grained soils (0.025 mm < PSD < 3.35 mm). The mentioned RCP is generated using a heuristic algorithm with merging applicable equations of soil science. For porous media modeling, MATLAB software is used and the predicted results by the optimal model for the parameters of porosity, pressure drop, and saturated hydraulic conductivity are compared with laboratory measurements. Experimental design is conducted by MINITAB and predicted coarse-grained soil structure by the model is compared with four sifted soils. The results of the sensitivity analysis showed that the porosity obtained from the model is strongly sensitive to the resolution factor and should be chosen with a sufficiently large amount (higher than 250). Results showed good consistency (up to 95%) between predicted porosity and only a 10% difference in pressure drop and permeability with observed measurements.

多孔介质结构的知识是多孔介质中流体流动的流体动力学研究的重要组成部分。为了研究包气带中土壤行为（作为粒状多孔介质）以及水和污染物的运动，适当估计土壤保水曲线（SWRC）和土壤导水率曲线（SHCC）具有关键作用，并且是最具挑战性的方法之一。土壤和水科学领域的研究人员和工程师的主题。SWCR 可以使用精确的粒度分布 (PSD) 函数来近似。在这项研究中，通过应用随机密堆积 (RCP) 方法作为预测和研究颗粒配置的一种令人鼓舞的方法，为粗粒土壤 (0.025 毫米 < PSD < 3.35 毫米) 开发了最佳粒度分布。提到的 RCP 是使用启发式算法生成的，并合并了土壤科学的适用方程。对于多孔介质建模，使用MATLAB软件，将孔隙度、压降和饱和导水率参数的最优模型的预测结果与实验室测量结果进行比较。MINITAB 进行了试验设计，将模型预测的粗粒土结构与四种筛分土进行了比较。敏感性分析结果表明，模型得到的孔隙度对分辨率因子非常敏感，应选择足够大的数值（大于250）。结果显示预测的孔隙度之间具有良好的一致性（高达 95%），而在观察到的测量结果中，压降和渗透率的差异仅为 10%。