JOURNAL OF HYDRAULICS
Year:2015 | Volume:9 | Issue:4|Papers Count:7

The present research indicates the ability of mesh-free natural element numerical method in simulation of free-surface flow and consistency of extended second order method of Van-Leer scheme on solving natural element method. The natural element method is based on natural neighbor interpolation. In this study, Sibson interpolation has been used. Nodal integration has been used due to inappropriateness of Gaussian point’s integration. The present method has been validated by two-dimensional heat transfer problem for the diffusion term validation and by one-dimensional dam break problem for the convective term evaluation. Simulation of free-surface flow has been analyzed on non-uniform bed in the channel with a bump. The results obtained are compared with analytical results. In addition, water flow inside the Parshall flume is simulated and results were compared with experimental interpretations. Results of modeling and analytical solutions via normalized root mean square error (NRMSE) for temperature problems, dam break and free-surface flow show NRMSE of less than 2% while 18% for Parshall flume test. In addition, the determination coefficient of numerical model for the first three tests are higher than 0.99 and 0.90 for the Parshall flume which suggests high capability of numerical model in simulation of free-surface flow.

Scouring downstream of overflow spillways is an important factor in the design of plunge pools and tail-pond dams. One of the methods to reduce scouring of falling jets is employing splitters on overflow spillway. Splitters cause the falling jet to expand and entrain air. As a result, the jet strength for scouring reduces. In the present work the scour hole extent and slopes located downstream of an overflow spillway with or without splitters is studied in a large hydraulic model with 6 m height. Experiments were conducted with different discharges and tail water depths. It was found that splitters reduce the scour hole slopes as well as the scour extent (width and length). Experiments show that with splitters the maximum width and length of the scour hole reduced by 17% and 20%, respectively. It was also concluded that with splitters, the ratio of scour hole length to its depth increased by 12% and the ratio of scour hole width to its depth increased by 14%. It was also observed that the scour hole slopes reduce in all experiments when splitters are present.

Scouring is one of the issues that is important in the design of various types of hydraulic structures. One of these structures is a stepped spillway. The aim of this study is to evaluate the similarity between the profiles of scour holes downstream of the stepped spillway. To do so, results of 67 experiments in two different experimental models were used; these results were in a wide range of particle Froude number, the stilling basin length, downstream depth, sediment particle size distribution and two different slopes of stepped spillway. In this study, 5 different lengths of stilling basin, from 37 cm to 120 cm, were used. Also, the range of utilized discharge was from 8.43 lit/s to 88.95 lit/s. Duration of the tests were selected to be 6-hours, 8 hours, 12 hours and 24 hours and, in total, 834 scour profiles and approximately 85000 points to were used derive required relations and to perform other analyzes. Based on the experiments results, the similarity between scour-holes located in downstream of stepped spillway was analyzed. In this process, two methods were used to make the profiles dimensionless. Based on statistical parameters, best method was selected and similarity of downstream profiles were shown. Relations that predict maximum scour depth, the distance from maximum depth to the end of the stilling basin and volume of transported sediments are developed based on the dimensionless parameters in different time periods. Furthermore, impact of different parameters on the geometry of scour holes were studied. When the Froude number increases, while all other parameters are constant, the dimensions of the scour-hole increases. When the stepped spillway slope increases, the dimensions of the scour-hole decreases and, moreover, when the tail water depth increases, while all other parameters are constant, a deeper hole with longer longitudinal distance forms. Overall, 8 cross section profiles under different conditions were taken. These profiles indicate transverse development of the scour-hole, in addition to longitudinal development.

The piers’ shape may be considered as one of the key factors in controlling the scour process around it, as this mainly changes the stream flow pattern around the structure. Furthermore, piers with variable section in depth may also be of interest in reducing the scour around them. The aim of this study was to evaluate the effectiveness of the piers with lateral boundaries fitted by velocity profile due to steady flow, in reducing local scour, when compared to the circular and conical shaped piers. The section of the piers was circular and the widths or diameters of these piers reduce from bed to water surface with a slope fitted by velocity profile. Due to increasing flow discharge close to the water surface, the flow which has maximum velocity in this region will then pass around the piers easier than the cylindrical pier and its velocity is reduced. Therefore, in using these piers, the dynamic pressure in upper regions decreases relative to that of the cylindrical pier and there will be a weaker down flow in front of pier which will consequently cause decreasing of the scour depth. The models tested in this study were three models with logarithmic profiles fitted by (d50=0.78, 1.1, 1.3 mm) and three models with conical profiles by (degrees, in which j is the angle of lateral slope). Then, the relative scour depth for these models compared with the cylindrical reference model and the maximum scour depth decreased with the rate of 55.9%, 65.4%, 73.2%, 38.5%, 47.2% and 52.7%, respectively. These tests have been performed in a wide flume at hydraulic research laboratory and under clear water conditions over12 hours using relatively uniform sediments particles. For the numerical models, the maximum scour depth around the piers was simulated using Flow-3D model. It was found that the numerical model results were in reasonably good agreement with the experimental data.

Spur-dikes are river cross structures attached to river banks which cause flow deflection. Optimization modeling of spur-dikes is composed of three modules: morphological, design and optimization module. Morphological module is applied to investigate river stability. Stable cross-section dimensions are computed by maximum sediment transport capacity of external hypothesis. Then, by applying appropriate design criteria for spur-dikes components, such as length, height, distance, number and scour hole depth, construction cost are minimized in design module. Finally, results obtained from morphological and design modules are formulated as a nonlinear multi objective optimization function by weighting method in optimization module to compute optimum dimensions of the spur-dikes. In other words, optimization module is a weighting combination of two objective functions of design and morphological module. The developed model computes variation of construction cost of super-dike with respect to sediment transport capacity of the canal, an indicative of river stability. By applying Zanjanrood river data near Sarcham Bridge, the model has been calibrated for weighting coefficient of 0.6 and 0.4 for cost and sediment transport capacity functions, respectively. Sensitivity analysis has also been carried out with respect to bed load and scour hole depth equations, which shows that they play an important role in determination of the super-dikes dimensions and construction cost. Among the bed load equations applied, Englund & Fredsoe’s have shown maximum sediment transport capacity and minimum cost while van Rijn’s, on the contrary, performed minimum sediment transport capacity and maximum cost.

The accurate and quantitative description of gravel bed roughness is essential in hydraulic studies. Traditional approach for characterization of random bed roughness relies on grain size distribution curve. However, in recent years, as an alternative approach, gravel bed is considered as a random field of bed elevations which varies point by point in longitudinal and transversal directions. In fact, in new approach, a more comprehensive description of the gravel bed is provided by the statistical analysis (e.g. correlations, spectra, structure functions) of bed elevations. In present laboratory study, the results of employing these two approaches to an artificial gravel bed are reported. Existing methods for estimation of grain size distribution curve (sieve analysis, image analysis and caliper measurements) are reviewed and their results are compared. The comparison shows that grain size measurement using caliper is the most suitable method. Moreover, comparison between characteristics of artificial gravel bed in present study and natural water-worked gravel bed in rivers, obtained from statistical analysis, shows that despite the simplicity of the employed method for creating our laboratory gravel bed, most of their characteristics are different. This observation highlights the importance of statistical characterization of roughness. Therefore, in laboratory study of gravel bed, it is recommended to precisely determine roughness characteristics. This will quantify the difference of natural river beds and artificial rough bed in experimental flumes.