Introduction
With population growth and the decline of water resources, optimal management of water resources, enhancement of hydraulic performance, and improvement of water distribution efficiency have become essential. Semi-circular radial gates require less force for operation due to their rotational movement around a vertical axis. On the other hand, Trapezoidal canals are the most common canal type used in irrigation and drainage networks due to their hydraulic and structural stability. Since Experimental studies are costly and time-consuming; therefore, numerical simulation is widely used to obtain reliable results for evaluating hydraulic behavior. Thus, the present study aims to numerically investigate the performance of a semi-circular rotating gate in a trapezoidal canal under free flow conditions using the Flow-3D software. The main objectives include evaluating turbulence models, analyzing stage-discharge relationships, identifying the inflection angle, and assessing energy losses for different wall slopes, transition lengths, and gate opening angles. The results of numerical model are validated by comparing simulation results with experimental data. If successfully validated, researchers can confidently use this numerical approach without costly experiments.
Methodology
sing galvanized steel sheets inside a rectangular flume, trapezoidal canals with bottom widths of 0.2, 0.3, and 0.4 m and corresponding side slopes of 0.44, 0.31, and 0.20 were constructed. To connect the canal to the semi-circular rotating gate, three gradual transition lengths of 0.6, 0.9, and 1.2 m were used.in the laboratory, Flow depth was measured by using a point gauge with an accuracy of 0.1 mm, discharge was measured with an electromagnetic flowmeter accurate to 0.01 lit/sec., and velocity measurements were conducted using a Velocity Profiler. Numerical modeling was performed using Flow-3D software. Gate opening angles ranged from 50° to 80° with 5° increments, and discharge rates ranged from 0.024 to 0.036 m³/s with 0.002 increments Mesh size was determined through a mesh convergence analysis. Using three turbulence models—standard k-ε, RNG, and LES—the water surface profile along the canal and the vertical velocity profile at the 2 m section were computed, and the best model for simulation was selected. Finally, using the numerical model results, the effects of variable parameters on water surface profiles, transverse profile variations, stage-discharge relationships, and energy loss were investigated and compared with experimental data.
Results and Discussion
Comparison of numerical and experimental results showed that all three turbulence models (standard k-ε, RNG k-ε, and LES) systematically underpredicted water depth and overpredicted velocity. This error arises from the isotropic Reynolds stress assumption in RANS models, which contradicts the highly anisotropic flow in the jet impingement zone. As wall slope decreases, flow moves more freely across the canal, increasing anisotropy. In the steep slope (0.44), flow converges at the canal centerline, reducing jet anisotropy. In the mild slope (0.20), jets move freely, creating a stronger impingement zone with higher anisotropy. LES, by directly resolving the three-dimensional flow structure, simulates this anisotropy better and performs superiorly for velocity profiles. However, RNG performs better for water surface profiles and energy loss due to its correction term in the dissipation rate equation.
Downstream of the gate, impingement of flows from both sides creates turbulence and a hydraulic jump (hump), whose location depends on the gate opening angle. As opening angle increases, the upstream-downstream depth difference decreases, and turbulence reduces. Wall slope significantly affects the flow field; as slope decreases from 0.44 to 0.20, cross-sectional area increases, flow becomes less concentrated, and impingement intensifies, increasing turbulence.
Canal slope dominates the effect of transition length. In longer transitions, flow becomes more uniform and fully developed, reducing downstream turbulence. In the steep slope (0.44), increasing transition length has little effect. Maximum energy loss occurs at L=0.6 m.
As bottom width increases (slope decreases), flow capacity decreases, and the stage-discharge curve slope increases, meaning small discharge changes cause significant depth changes. At small opening angles, the gate acts like an orifice with very low flow efficiency. As opening angle increases, outlet area and flow capacity increase. The results indicate a change in the hydraulic behavior of the flow within the 60-65degree range.



Conclusion
Comparison of numerical and experimental results showed that no single turbulence model is superior in all aspects. The appropriate model should be selected based on the output required. Flow behavior strongly depends on side slope. In the mild slope (0.20), jets impinge with maximum energy, creating a highly turbulent zone with non-uniform Reynolds stress distribution. As gate opening angle increases, the stage-discharge curve slope decreases. At small angles, the gate acts as an orifice. Numerical model predictions showed good agreement with experimental data.