Introduction: Sediment transport is one of the most basic and important characteristics in river hydraulics and bed morphology. The prediction of sediment transport path in rivers and also cannels is absolutely complicated, and mostly conducted with semi-empirical methods. In such cases, the Lagrangian method is essential for exploring the physics of individual sediment particles. The investigation of the flow pattern in the compound open-channel originated in 1960s and followed by the exploration of turbulence structures of overbank flows. However, studies on the characteristics and processes of sediment transport in the compound channels are rarely conducted. For completion this gap, in this experimental study, the rolling and sliding motions of individual bed particle in the floodplain of a rectangular compound open-channel have been experimentally investigated. Specifically, the mechanical parameters of particle motions such as velocity and acceleration are investigated. In this regard, different statistical distributions, especially Gaussian or normal distribution, are employed to introduce the properties of bed sediment motions in the floodplain. Methodology: The experiments were conducted in the hydraulic laboratory of Tarbiat Modarres University in a straight open channel with length of 10 m, width of 1 m and height of 0. 7 m (Fig. 1). The laboratory flume is a wide rectangular channel with a compound section (Fig. 2), where the side wall and bottom of the channel are made of glass. The main channel is 0. 4 m wide and the floodplain is 0. 6 m wide. To control the water depth, an adjustable weir was used at the end of channel. The discharge at the inlet of the channel was controlled using a regulating valve downstream of pump and measured by an electromagnetic flow-meter. The hydraulic conditions of the experiments are summarized in Table 1. According to the calculations, the Reynolds and Froude numbers are respectively 28000 and 0. 34. Therefore, the flow in the compound channel of the present study is turbulent and subcritical. The flow depths in the floodplain and main channel are 5 and 20 cm, respectively. To capture high quality images from bed particle motions in short intervals, a camera with the speed of 24 frames per second and FullHD resolution was used (Fig. 3). To improve the quality of the images, the floodplain and main channel bottoms were coated with black color in the measurement zone. Moreover, for detection of the particle trajectories, the measurement zone was regularly meshed by the perpendicular lines with the distance of 10 cm. Several projectors were applied at different angles for illumination of the measuring plane. The spherical bed particle characteristics of the present study are mentioned in Table 2. Particle tracking were conducted at the distances of 5, 20, 40, and 50 cm from the floodplain side wall (Fig. 4), and repeated about 20 times for each one. Results and discussion: Chi-Squared test were used to determine the appropriate distribution to describe the longitudinal and transversal velocity and acceleration of individual particles (Fig. 5). Also, skewness and kurtosis of the data are employed to investigate the fitness of velocity and acceleration data to the normal distribution (Eqs. 2 and 3). In the case of sediment release at 20 cm from the floodplain side wall, the skewness values for the particle longitudinal and transversal velocities are always close to zero and their kurtosis values are close to 3, . This indicates that the particle longitudinal and transversal velocities follow the normal distribution. However, kurtosis of longitudinal acceleration diverges from 3, and consequently, it does not follow normal distribution (Table 3). The averaged longitudinal and transversal velocities of the sediment particles increase, approaching to the interaction zone (Fig. 6). Also, the standard deviation of longitudinal and transversal velocity and acceleration values increase with the increase of distance from the floodplain side wall (Fig. 7 and 8). Kurtosis of streamwise and spanwise velocity and acceleration of sediment particles increase far from floodplain side wall (Fig. 9), duo to the uniformity of particle motions in the interaction zone. The linear relationship between the average particle velocity and flow shear velocity indicates that there is a good agreement between the results of the present study and previous researches. Conclusion: The results of this study show that the sreamwise and lateral velocity and spanwise acceleration histograms of spherical particles in the floodplain far from the interaction zone, could be fitted to the normal distribution. While the kurtosis of histograms increases considerably, approaching to the junction. The histogram of streamwise acceleration does not fitted by the normal distribution. The histogram kurtosis of velocity and acceleration is enhanced approaching the interaction zone.

Prediction of flow through the compound open channel is one of the main problems in the field of hydraulic engineering. One of the main parameter related to the flow properties in the compound open channel is Shear Stress. The shear stress is because of difference of velocities between the main channel and floodplains. The Shear Stress causes of turbulence and vortex creation on the border of main channel and floodplains. The difference between the roughness of main channel and floodplains intensities the shear stress in the border zone and also decreases total discharge. In this paper, the discharge of flow in compound open channels was predicted using group method of data handling technique. To this end, related dataset were collected from literature. Involved parameters in modeling are relative hydraulic depth ( ), relative hydraulic radius ( ), and relative roughness ( ) and relative area ( ). To compare the performance of GMDH with other types of soft computing methods, the MLPNN as most well-known soft computing technique is developed as well. Results indicate that the GMDH model with coefficient of determination 0. 91 and root means square error 0. 057 is more accurate than the MLPNN. Reviewing the structure of developed GMDH model shows that and are the most effective parameters on prediction of discharge in compound open channels.

Differences in the flow properties in the main channel and flood plains causes, mass and momentum tensions between the both sections. The non-prismatic compound open channel cross section intensifies the mass and momentum transferring between the main channel and floodplains and has significant effect on the flow properties through the compound open channel.In this study the flow properties in the heterogeneous Roughness Non-Prismatic Compound Open Channel was assessed using the numerical and physical modeling. The physical modeling was conducted in the hydraulic laboratory center of Tehran University and numerical modeling was carried out using the Flow-3D as famous computation fluid dynamic tool (CFD).The results indicates that the Flow-3D is an effective tool for modeling the flow in the heterogeneous roughness non-prismatic compound open channel. During the CFD modeling it was found that the RNG turbulence model is more precise for simulation and modeling the flow properties. The results show that the heterogeneous roughness has significant effect on the flow characteristics such as velocity distribution and share stress gradient.

A compound open channel is composed of the main channel and flood plain. In experiments with compound open channel conducted to ensure that the flow is uniform and fully developed, it is necessary to study the distribution of discharge in the main channel and flood plains. The purpose of the present study is to investigate the effects of channel inlet condition on flow uniformity by considering distribution of discharge in channels with length to flood plain width ratio lower than 35 (needed for fully developed flow condition) by analyzing the flow field and turbulence parameters. For this purpose, particle image velocimetry method has been used in a rectangular compound open channel. To provide correct measurement of secondary velocities, using a non-intrusive method such as particle image velocimetry is completely essential. The results of this study show that in short compound channels with the same screen installed in the main channel and flood plain, there is significant mass transfer from the flood plain to the main channel until the end of the channel length. It was found that in this case, a considerable downfall occurs for the maximum velocity position in the main channel. However, with a supplementary screen installed in the flood plain, in addition to the typical screen, expected conditions are established similar to the fully developed compound channels. In this condition, the time-averaged streamwise velocities vary considerably in the flood plain along the spanwise direction. On the other hand, in short compound channels with the same screen installed in main channel and flood plain, the streamwise velocities do not change significantly along the flood plain width duo to the imperfect interaction of main channel and flood plain. These observations express that to provide correct distribution of discharge, a supplementary screen should be installed in the flood plain of the compound channel.

Modeling of flow through the compound open channel is one of the main problems in the field of hydraulic engineering. One of the main parameter related to the flow properties in the compound open channel is Shear Stress. The shear stress is because of difference of velocities between the main channel and floodplains. The Shear Stress causes of turbulence and vortex creation on the border of main channel and floodplains. The difference between the roughness of main channel and floodplains intensities the shear stress in the border zone. In this investigation using the physical and numerical modeling the flow properties in the heterogeneous roughness prismatic compound open channel was studied. The study was carried out in the hydraulic laboratory of Tehran University and numerical modeling was conducted using the Flow3D as famous computational fluid dynamic (CFD) tool. The results indicated that the Flow3D software has high ability for modeling the flow characteristics in heterogeneous roughness prismatic compound open channel and the RNG turbulence mode is suitable for modeling the vortex on the border of both sections.

Introduction: A compound channel consists of one main channel with a deeper flow in the middle and one or two floodplains around the main channel with lower flow depth. The difference between velocity in the main channel and on the floodplains in compound channels creates a strong shear layer at the interface between the main channel and floodplains. Also, because of the three-dimensional (3D) structure of flow, the investigation of flow characteristics in compound channels is completely complicated. In non-prismatic compound channels, due to the mass exchange between subsections, the study of flow is more complex. Therefore, the prediction of flow behavior in the non-prismatic compound channel is an important subject for river and hydraulic engineers. The skewed compound channel is one kind of non-prismatic compound channels. In compound channel with skewed floodplains, one of the floodplains is divergent and the other is convergent. The flow patterns in skewed compound channels have been studied experimentally by many researchers (James and Brown, 1977; Jasem, 1990; Elliott, 1990; Ervine and Jasem, 1995; Chlebek, 2009; Bousmar et al., 2012). However, numerical studies on flow characteristics in skewed compound channels were rarely performed. In this research, the velocity, boundary shear stress distributions, secondary current circulation, and water surface profile in a compound channel with skewed floodplains have been numerically investigated using the Computational Fluid Dynamics (CFD) and two turbulence models of the RNG and LES. Methodology: In the present research, modeled compound channel is similar to the experimental channel used by Chlebek (2009) at the hydraulic laboratory of Birmingham University, Department of Civil Engineering. The experimental studies were performed in a straight flume of 17 m long, 1. 198 m wide, 0. 4 m deep, and with an average bed slope of 2. 003×10-3. The PVC material was used to make compound cross-section. A rectangular main channel of 0. 398 m wide and 0. 05 m deep in the middle, and two floodplains with 0. 4 m wide around the main channel (Fig. 2). The skewed compound channel was made by isolated floodplains using L-shaped aluminum profiles. Experiments were conducted at the skew angle of 3. 81° and four relative depths of 0. 205, 0. 313, 0. 415, and 0. 514. The lateral distributions of depth-averaged velocity and boundary shear stress were measured at six sections along the skewed part of the flume (see Fig. 3), using a Novar Nixon miniature propeller current meter and Preston tube of 4. 77 mm diameter, respectively. For numerical simulations of the flow field in the skewed compound channel, the FLOW-3D computational software was used. Also, the renormalization group (RNG) and Large Eddy Simulation (LES) turbulence models were selected as turbulence closure. Two mesh blocks were utilized for gridding, mesh block 1 by coarser mesh size at the upstream of the skewed portion of the channel, and mesh block 2 by smaller mesh size for skewed part (Fig. 5). The flow field is numerically simulated by three computational meshes (fine, medium, and coarse mesh size). Details of gridding for different computational meshes are summarized in Table 2. Finally, the medium mesh by 1653498 cells was selected. For boundary conditions, using volume flow rate condition for inlet, outflow condition for the outlet, symmetry condition for water surface area and the interface of two mesh blocks, and wall condition for lateral boundaries and floor (see Fig. 8 and Table 3). Results and Discussion: The results of the numerical simulations show that the RNG turbulence model, can predict the depth-averaged velocity and boundary shear stress distributions in the skewed compound channel fairly well (Figs. 9 and 10). In addition, in the skewed compound channel, the mean velocity and boundary shear stress on the diverging floodplain is more than converging floodplain at the same section. The longitudinal discharge distribution on floodplains of the skewed compound channel is linear, and the numerical modeling can compute those values very well (Figs. 11 and 12). By moving along the skewed part of the flume, the regions with higher velocity move toward the diverging floodplain. Also, the position of the maximum velocity, instead of the main channel centerline, move to the interface between the main channel and diverging floodplain (see Figs. 13 and 14). The lateral flow that leaves the converging floodplain, plunging into the main channel flow, creates a secondary flow circulation in the main channel and near the converging floodplain. Also, as moving along the flume and get close to the end of the skewed portion, this secondary flow becomes stronger (Figs. 15 and 16). Regarding the water surface profile in the skewed compound channel, two turbulence models can predict the water depth along the channel fairly well, especially the RNG turbulence model (Fig. 17). In addition, the error analysis by using experimental data and numerical results are investigated. For error analysis, Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE), and the coefficient of determination (R 2 ) were calculated by using the equations of (12) to (15), respectively. The computational errors between the results of numerical simulation and experimental data are presented in Table 5 and are showed in Figs. 18 and 19. Conclusion: In this research, the flow field in a compound channel with skewed floodplains has been numerically simulated. The FLOW-3D software and two turbulence models of the RNG and the LES were used to model the depth-averaged velocity, boundary shear stress distributions, and discharge distribution at different sections along the skewed compound channel. The results of simulations indicated that compared to the LES turbulence model, the RNG turbulence model are able to predict the velocity and bed shear stress distributions quite well especially at the first half of the skewed portion. Also, by increasing the flow depth, the accuracy of numerical modeling for prediction of the velocity and bed shear stress increase, while for the water surface profile decreases (see Fig. 18).

Water level determination during a flood is always a challenging task for river engineers. During the flood, river channel becomes compound consisting of the main river channel which carries low flows and floodplains that carry overbank flows.  Flow velocity and structures are affected by vegetation, the degree to which depends on vegetation density, flexibility, type, and whether it is in a submerged or emergent condition. Water surface modeling help for the study of flood waves, water level calculation during flood, stage discharge relation, design of water work structures. This work develops a model which can be used to simulate water surface profile in compound channel with vegetated floodplains with various vegetation covers. To predict the water surface, experiments have been conducted in the laboratory for different hydraulic conditions. It can be seen from the results that the trend of stage-discharge relationships is found to be an exponential function giving a high value of R2. A multivariable regression model (MRM) has been developed to predict the water surface profile for such channels. The dependency of water surface profiles on four different non-dimensional parameters such as canopy arrangement, canopy density, relative depth and relative distance are analyzed. Using the relevant experimental data, non-linear regression has been performed. The results obtained from the present water surface profile model shows good agreement with the observed data.

Secondary currents are important mechanisms in open channels, having major contribution in flow field and its corresponding parameters including boundary shear stress and depth-averaged velocity. Compound open channels involve extra plan form vortices in the flow field. Curved open channels generate especial vortices due to the effects of centrifugal force. Precise modeling of secondary currents in curves and meanders is a very important issue in practical applications. Using open source ``OpenFOAM'' software, the flow field in a meandering channel with compound cross section is simulated herein. Reynolds averaged Navier-Stocks equations (RANS) are solved. Applying appropriate boundary conditions over the free-surface, the simpleFoam solver has been used to model the two-phase air-water flow interface, assuming a steady flow condition, and a symmetry boundary condition. The experimental data from FCF belonging to University of Birmingham is selected for verification and validation of the present numerical results. Two turbulent models of Realizable k− varepsilon and SST k− omega are applied. Lateral velocity profiles at the cross sections indicate that at each wave-length of a meander, a vortex forms in the main channel at the apex, directing towards the outer bank near the bed and towards the inner bank near the water free-surface. The secondary current patterns, achieved for curved compound open channels differ from those of the simple channels. This is partly due to the interaction of shear stresses occurring at the interfaces between the main channel and the floodplains. Deviation of paths of the main channel and floodplains, downstream of each apex, results in entering the flow from inner bank to the main channel and exiting the flow from the main channel to the outer bank. These flow patterns shift the flow from inner-to the outer bank, downstream of each apex. Therefore, a helicoidally secondary current pattern forms, growing in size and strength farther downstream of the apex region.