Experimental and Numerical Investigation of Un-Steady Non-Darcy Flow in Rockfill Materials (RFM)

• Published in: Water resources 2023-12, Vol.50 (Suppl 2), p.S164-S171
• Main Authors: Hajikazemian, H., Bazargan, J., Shokri, M., Safarian, M., Norouzi, H.
• Format: Article
• Language: English
• Subjects: Aquatic Pollution; Coefficients; Computation; Computer applications; data collection; Earth and Environmental Science; Earth Sciences; equations; Flow velocity; Hatching; Hydrogeology; Hydrology/Water Resources; Hydrophysical Processes; Numerical analysis; Porous materials; Porous media; Rockfill; Sensitivity analysis; Surface water; transient flow; Unsteady flow; Waste Water Technology; Water depth; Water Management; Water Pollution Control
• ISSN: 0097-8078; 1608-344X
• DOI: 10.1134/S0097807821100791

In this work, we studied the behavior of the unsteady flow in rockfill materials (i.e., grained porous media) both experimentally and numerically. The unsteady flow was created by moving up and down a hatch installed at the end of the channel. Experimental data were collected and stochastically analyzed, and Forchheimer relation coefficients were accurately calculated. Then, the Saint Venant equations were considered as governing relations to analyze unsteady flow. In this study, we considered all terms of Saint Venant equations contrary to previous studies where some terms have been excluded from the analysis. For scrutinizing responses of numerical calculations, a sensitivity analysis was conducted based on and . By using binomial and power relations (separately), the values of depth and velocity of flow were computed and observed surface water profiles obtained. The comparison of the results revealed the high accuracy of computations so that the maximum computational error in moving up and down of the hatch was computed to be 3.4 and 2.9%, respectively. The results indicated the superiority of binomial equations to power, so that the maximum value of computational errors in binomial relations shows, on average, 22% relative improvement than power equations.