مجله هیدرولیک
سال پانزدهم شماره 4 (زمستان 1399)

Large landslides can cause overtopping and consequently demolish the dams and other substructures and facilities. The landslide stabilization is very costly due to their large size and considerable extent. Hence proper estimation of the wave height because of sliding into the reservoirs in order to determine the risk of overtop is inevitable. In this study, 3D simulation of the SM5 sliding landslide into the upper reservoir of the Siyah Bisheh is conducted to calculate the height of the waves generated by this phenomenon.

For landslide modeling, the assumption of rigid mass is assumed and mass motion is considered as a combination of transitional and rotational motion. First, Heinrich's (1992) laboratory model was used to evaluate the performance of the Flow-3D numerical model (calibration and validation). For this purpose, the sliding mass was introduced into the reservoir by prescribed motion and the changes in water level at different points were compared with the experimental results. The results showed that there is an appropriate matching between the experimental and numerical results. The purpose of this comparison was to evaluate the accuracy of the software used to estimate water level. Although in some cases, the trend of changes in water level is significantly different from laboratory results, the maximum level obtained in numerical model is in relevant agreement with laboratory results. In the numerical simulation of the mass movement in the Siyah-Bisheh Dam, the mass range and the shape of the slide circle are first determined. The mass has a volume about 425,000 cubic meters and is about 100 meters long. The slide radius is estimated to be 200 meters. About 250,000 cubic meters of this mass lies beneath the reservoir water level, which 150,000 cubic meters moves during landslide. In this case, the center of mass moved 20 meters downwards and can generate a velocity between 1 to 10 meters per second. For the modeling of motion, different scenarios are considered based on the mass movement velocity. Due to the low width of the river at the mass location (about 60 m), the mass movement is limited at this distance so all masses cannot enter to reservoir. Topography of reservoirs with 1/ 2000 scale were used to model reservoir and SM5 mass. Because of the narrow width of the valley, the mass hits the opposite wall and stops. As mentioned before, mass movement is considered a set of rotations and translations. The main reason of using this type of movement is the lack of ability to consider mass deformation. In terms of mesh size, meshes of 5 and 10 m in plan and 1 and 2 m in height are used. The results of convergence test show that there is no significant difference between the meshes of 5 m in the plan and 2 m in the depth and finer one. Different scenarios with various velocity of mass (velocity of 1, 3, 6 and 10 m / s) are considered for the simulation process. The mass is assumed to have reached maximum speed in a short time and stopped shortly at the end of the opposite wall. Results are presented for the 5 specified points in the reservoir with an appropriate distribution on its surface.

The results are presented for 90 seconds after the mass enters the reservoir and it has been attempted to take into account the impact of the distance and time when the peak occurred. The wave height near the mass reaches to 10 meters where the mass has 10 meters per second and reaches to around 3 meters as it departs from the entry point. The maximum wave height near the dam site has been obtained around 2.5 m. According to the laboratory results, the wave caused by the landslide moves in the direction of mass entry into the reservoir. The mass direction is perpendicular to the river and parallel to the dam axis and it is expected that the generated wave will hit the opposite wall. The generated wave due to landslide collides to the opposite bank and dispersed. As a result, the height of the generated wave is reduced and therefore the possibility of overtopping falls dramatically. Based on the results, it can be said that the wave height will not exceed 2.5 m near the dam body. However, the maximum wave height produced in the reservoir exceeds 10 m at high velocities. At the end, the surface wave height due to landslide has been calculated using the issue number 53 of Iranian Commission on Large Dams (IRCOLD). In this calculation, the slide mass is estimated to be 150,000 cubic meters and the mass velocity is 13 m/s.

According to the empirical tables and relationships, the wave height is obtained at 400 and 800 m far from the dam body at 2.5 and 1 m respectively. This value is compatible with the results obtained from the numerical model. Keywords: Landslide, 3D numerical modeling, wave height, overtopping, Siyah Bishe

In this research work, the maximized transmitted torque due to the impulse from flowing water in an open channel has been studied for five types of cylindrical turbines to find the best water turbine in terms of maximum produced electrical energy. For this purpose, using numerical finite volume method, a set of turbine and blades, consisting of a 3-dimensional cylindrical water turbine of equal diameter and length (1m), with five different blade configurations, has been simulated in a 10 m long and 3 m wide rectangular open channel with no inclination, subjected to a water flow of 2 m/s velocity. Considering the weight of various elements, the set of turbine and blades has been designed so that it remains floating in the channel at various immersion depths. Furthermore, with change in flow depth, the immersion depth remains constant. Considering the magnitude of the flowing water impulse in the channel, the corresponding torque transmitted from the water to the blades of the five types of turbines was determined and the maximum torque value was obtained.

In the present research, five types of blades, attached to a hollow cylindrical turbine of 1 m length and 1 m diameter, have been used. The turbine floats on water at a particular depth in an open channel. The water speed in the open channel determines the torque due to the impulse from the flowing water. Considering the various blades, the resultant torque has been studied numerically using two-phase flow finite volume method. The material for the construction of the cylindrical turbine and the blades is dense polyethylene having a density of with 950 kg/m3. Moreover, the fluids considered in the finite volume numerical computations are water with a density of 998 kg/m3, and air with a density of 1 kg/m3 at a constant temperature of 20° Celsius. The hollow turbine cylinder has a volume of 0.78 m3 and is filled with air. The three-dimensional flow channel, in which the turbine is placed, is a rectangular concrete channel of 10 m length and 3 m width, through which water flow at a depth of 35 cm. To avoid the effect of surrounding walls on the transfer of the flow impulse to the turbine, width of the channel has been considered slightly oversized. The roughness values for the bottom surface of the concrete channel and the turbine walls and the blade set is 1 mm and 0.01 mm, respectively. The depth of the channel is constant at 1 m, which is equivalent to the average depth of most city open channels.

The difference between type 1 and type 2 turbines is in the blade’s angle along the turbine rotational axis. As a result, the produced torque by type 1 turbine is more than that of type 2. On one hand, increase in contact area between the blades and the flowing water in the channel results in higher torques. On the other hand, the angular shape of the increases the slip between the flow and the blades, which reduces the conversion of kinetic energy into static energy. Ultimately, the result of the above two phenomena in type 2 turbine is a reduction in the produced torque to about 15 N/m. In type 3 turbine, an increase in the produced torque was achieved through the increase in the number and shape of the turbine blades. Hence, the implementation of the aforementioned changes relative to type 2 turbine resulted in an increase in the produced torque to about 56 N/m. Therefore, increase in the number of blades and change in the blade shape in this type of turbine compensated for the effects of blade angle elongation in type 2 turbine. Furthermore, in type 4 turbine, the internal diameter of the blades was reduced, while the number of blades was increased, blades distribution angle was changed from 45° to a straight configuration, and the contact area also decreased. Consequently, the amount of flow slip along the blades also decreased. Specifically, the result of all the above mentioned changes in type 4 turbine was to reduce the produced torque to about 14 N/m compared to type 3 turbine. Therefore, the combined effect of reduction in the contact area and reduced internal diameter of the turbine blades is more dominant than the combined effect of increased number of blades and reduced flow slip on the blades. In type 5 turbine, number of blades was reduced by 10 and the blades internal diameter was tripled relative to type 4 turbine, which resulted in a significant increase in the produced torque. Therefore, type 5 turbine, as a floating turbine, may be recommended for production of electric energy in open channels.

Considering the results of the calculations, type 5 turbine with 21 semicircular shape blades, 11 cm in external diameter and 10.5 cm internal diameter, has a higher capacity to produce more torque compared to other types of turbine studied. The factors affecting the final produced torque include the contact area between the blades and the flowing water in the channel, the length of the blades along the turbine axis, the extent of slip of flow when facing the turbine blades, and the number of blades. The produced torque by type 5 turbine is 458.96 N/m, which is the highest among the turbine types studied in this research.

Vegetation has traditionally been viewed as a nuisance and obstruction to channel flow by increasing flow resistance and water depth. However, in recent years, vegetation has become a major component of erosion control and stream restoration. Most of research efforts focus on describing vegetation roughness , determining drag coefficients and empirical formulas for resistance under various vegetation configurations. While the development of experimental solutions for vegetative resistance is important, it is also important to understand the detailed characteristics of flow through vegetation. Yang et al.(2007) conducted flume experiments with different types of vegetation, and found that, in the cases of non vegetated floodplains, all measured streamwise velocity distributions followed the logarithmic distribution, but for vegetated floodplains, they followed an S-shaped profile Nezu and Sanju(2008) have investigated turbulence structures and coherent motion in vegetated canopy open-channel flows. They divided the whole flow region into three sub-zones, i.e., the emergent zone, the mixing-layer zone and the log-law zone. In the present study, some experiments were undertaken herein under different conditions to elucidate the flow structure. The main focus is to examine how the vertical velocities, are affected by simulated vegetation arranged in emergent and submerged conditions. In addition, the effect of dowel density, configuration, and relative depth are examined.

The experiments were conducted in a fixed bed rectangular flume, 9 m long and 0.6 m high and 0.8 m wide. The slope of bed flume was 12 ×10-5. The main channel and floodplain had widths of 24 and 28 cm, respectively, and the main channel had a side slope, s, of 0. The bankfull height, h, was 6 cm. Vegetation were simulated by wooden dowels. The wooden dowels were 140 mm tall and 7 mm in diameter. The dowels were attached to a PVC sheet bolted to the bottom of the flood plain in linear and staggered arrangement. The spacing of the dowels varies from 2.5-10 cm in both lateral and streamwise directions forming stem density of 0.41, 1.64%, 6.04%. The flume was operated under a uniform flow condition, and measurements of discharge, point velocity and flow depth were taken. Flow depths were measured by means of a pointer gauge, discharges were measured by a digital flowmeter, installed upstream of the channel, and a micro propeller current meter were used to velocity measurements. Within the measurement cross section, located at 5.6 m, the authors arranged ten verticals, where the lateral values of y from the first vertical to the last were 0, 4, 8, 12, 12.2, 26 and 34 cm. When the vertical distance from the measurement point to the bed was less than 175 mm, the measurement interval was 10 mm and 5mm in the main channel and floodplain, respectively.

The experimental results are presented in three parts, flow through non-vegetated floodplain first, flow through emergent vegetation second, followed by the submerged case. The effects of density and dowel configuration are included in each of the sections. Each section ends with a discussion on the effects of rigid dowels on logarithmic profile. In the cases of nonvegetated floodplains, all measured streamwise velocity distributions followed the logarithmic distribution, but for vegetated floodplains, they followed an S-shaped profile. It is seen that after implanting the vegetation over the floodplain, the velocity over the floodplain decreases whereas it increases in the main channel. Also, as the vegetation density, λ, increases, velocity increases in the main channel and decreases in the floodplain. In the presence of emergent vegetation on floodplain, logarithmic profile does not exist even in the main channel, however it seems that the formation of the S-shaped profile in the main channel is under the bankfull height and above the bankfull height the vertical velocity profile takes on a logarithmic profile again. Under Submerged flow conditions, the velocity characteristics in all locations above the dowel array are well illustrated by the semi-logarithmic expression that has a slip velocity initially near the inflection point. On the basis of the present experimental results, the whole flow region is divided into the following three sub-zones: (1) Emergent zone (0 ≤ z ≤ hp), (2) Mixing-layer zone (hp < z ≤ hlog), (3) Log-law zone (hlog < z ≤ H). In the present study, hp was equal to 0.2 H and hlog was equal to 0.5 H. In the emergent zone (0 ≤ z ≤ hp) the velocity is almost constant due to strong wake effects of vegetation stems although it may behave slightly in a counter-gradient fashion. In the second zone (hp ≤ z ≤ hlog), the vertical velocity profile are similar in both submerged and emergent conditions, and the effect of bed roughness is completely eliminated and the velocity gradients are reduced and almost fixed. The velocity in the third zone (hlog < z ≤ H) is significantly higher than the velocity in the second zone. In the log-law zone (hlog < z ≤ H), the log-law of velocity distribution for rough beds is reasonably applied even to vegetated flows. Comparison the longitudinal velocity profiles for linear and staggered dowel arrangements indicates an increase in the resistance due to the linear arrangement compared to the staggered arrangement.

In the cases of non vegetated floodplains, all measured streamwise velocity distributions followed the logarithmic distribution, but for vegetated floodplains, they followed an S-shaped profile. However, in the main channel, higher than the bankfull height the velocity profile is logarithmic. The results shows that as the vegetation density, λ, increases, the velocity increases in the main channel and decreases in the floodplain. Linear arrangement resulted higher resistance compared to staggered vegetation arrangement. The velocity profile at all locations above the dowel array are very well represented by the following semi logarithmic expression. In fully submerged vegetation, the whole flow region was divided into three sub-zones, i.e., the emergent zone, (0≤z≤hp) the mixing-layer zone (hp < z≤hlog), and the log-law zone(hlog<z≤H). In the present study, hp was equal to 0.2 H and hlog was equal to 0.5 H.

Dams have played an important role in the development of human civilization, the simplest of which is the provision of water resources in agriculture, industry and drinking. The share of earthen and gravel dams, which often have tunnel overflows and shoots, is significant. In dams with this type of overflow, increasing the height of the dam increases the flow velocity on the shot and increases the probability of cavitation. To protect hydraulic structures such as overflows, shoots, and lower discharges from cavitation damage, some air is typically added to the flow in areas with a cavitation index below the critical value. By using aerators, erosions that occur due to cavitation on overflow surfaces can be prevented. Aerators are usually installed on the floor and sometimes on the side walls of the overflow, separating the high-velocity currents from the surface of the overflow and preventing erosion at the rigid boundaries by artificially introducing air into the flow.

Most air inlet and outlet experiments focus on the average concentration of air in the stream and require the measurement of the amount and manner of air out of the stream. Computational fluid dynamics is a relatively new method and a review of studies in numerical modeling of overflows shows that the use of this tool as a research tool in research institutes began and gradually accepted by the hydraulic engineering community. Using computational fluid dynamics alongside physical models is a good way to reduce costs and save time. Due to the high accuracy of this method in determining the jump length of the flow jet, its results can be used to determine the geometric parameters of aerators such as the width of the air distribution duct. As the slope of the shot increases, the entry of air into the stream increases and also the changes in air concentration decrease, so the need for the presence of aerator decreases. Therefore, in the present study, using Fisher (2007) laboratory data to numerically simulate the flow through the aerator, the changes in air concentration along the shoot bed have been investigated. For this purpose, FLUENT software was used to model the two-phase air-water flow and the length of the flow jet jump was used as an important and effective factor in the entry of air into the flow. . Although aerators have been proposed since 1970’s but today there is no any reliable design guideline for determinate aerator spacing.

By determining the trend of changes in bed air concentration, the distance between two aerators can be determined. The air in the stream causes the stream to condense and dampens the shocks caused by the explosion and bursting of the bubbles, thus reducing the damage caused by cavitation; On the other hand, if more than necessary to prevent cavitation, air enters the stream, causing the flow to become bulky, and higher walls should be considered for the shot, which is not economically appropriate. Therefore, it is important to determine the minimum air concentration required to prevent cavitation damage. Determining the location of the second aerator can be determined according to the minimum required air concentration and the length of the flow jet for the height and landing number upstream of the first aerator. And the ramp angle increases and decreases as the water level above the aerator increases. Determining how the air concentration changes downstream of the shot aerator is important for calculating the distance of the aerators from each other, and FLUENT models the process of these changes well. The comparison between numerical and laboratory results shows a very good agreement between the laboratory and numerical model. Finally, a relation for the distribution of air concentration in the substrate was presented, which has a good fit with laboratory data. Due to the importance of the point of impact of the current to the shooting bed (sudden outflow of air due to the impact), the point was used as a reference for calculations.

The comparison between numerical and laboratory results shows a very good agreement between the laboratory and numerical model. In general, the results showed that the air concentration of the lower bed of the aerators increases with increasing landing number, ramp height, step height and ramp angle, and decreases with increasing water height upstream of the aerator, and increases with increasing slope of the air flow and also changes. The air concentration decreases, so the need for aeration is reduced. Keywords Minimum air concentration, Cavitation, Computational Fluid Dynamics, Overflow, Aeration.

Labyrinth weirs are among the hydraulic structures that are constructed to regulate the water level and control the flow in canals, rivers and reservoirs of dams. This structure is designed to transmit large currents in low heads by increasing the effective length of the overflow crown. The crest axis of this type of weirs is indirect and in showing the horizontal surface, the overflow is composed of interconnected walls. Labyrinth weirs are repeated with triangular, trapezoidal, rectangular and arc geometries alternating in flow width. The main criterion in the labyrinth design of the weirs was to increase the flow transmission capacity on the overflow with a fixed canopy and for a certain height of the water level upstream of the weir. In this study, 145 laboratory data were used, including four types of arc labyrinth weirs with different arc radius and tangent lengths. The discharge efficiency was introduced as a dimensionless parameter  using dimensional analysis. The relationships between  and the critical depth of flow (y_C/P) were obtained using graphs for two different types, R2 = 0.983 and R2 = 0.998. Then a graph to y_(C_Lt )/P and H_t/P a relation to calculate the critical depth were presented. The results showed that the calculated flow rate obtained using yc is consistent with laboratory values so that the relationship between them is R2 = 0.982.Labyrinth weirs are among the hydraulic structures that are constructed to regulate the water level and control the flow in canals, rivers and reservoirs of dams. This structure is designed to transmit large currents in low heads by increasing the effective length of the overflow crown. The crest axis of this type of weirs is indirect and in showing the horizontal surface, the overflow is composed of interconnected walls. Labyrinth weirs are repeated with triangular, trapezoidal, rectangular and arc geometries alternating in flow width. The main criterion in the labyrinth design of the weirs was to increase the flow transmission capacity on the overflow with a fixed canopy and for a certain height of the water level upstream of the weir. In this study, 145 laboratory data were used, including four types of arc labyrinth weirs with different arc radius and tangent lengths. The discharge efficiency was introduced as a dimensionless parameter  using dimensional analysis. The relationships between  and the critical depth of flow (y_C/P) were obtained using graphs for two different types, R2 = 0.983 and R2 = 0.998. Then a graph to y_(C_Lt )/P and H_t/P a relation to calculate the critical depth were presented. The results showed that the calculated flow rate obtained using yc is consistent with laboratory values so that the relationship between them is R2 = 0.982.Labyrinth weirs are among the hydraulic structures that are constructed to regulate the water level and control the flow in canals, rivers and reservoirs of dams. This structure is designed to transmit large currents in low heads by increasing the effective length of the overflow crown. The crest axis of this type of weirs is indirect and in showing the horizontal surface, the overflow is composed of interconnected walls. Labyrinth weirs are repeated with triangular, trapezoidal, rectangular and arc geometries alternating in flow width. The main criterion in the labyrinth design of the weirs was to increase the flow transmission capacity on the overflow with a fixed canopy and for a certain height of the water level upstream of the weir. In this study, 145 laboratory data were used, including four types of arc labyrinth weirs with different arc radius and tangent lengths. The discharge efficiency was introduced as a dimensionless parameter  using dimensional analysis. The relationships between  and the critical depth of flow (y_C/P) were obtained using graphs for two different types, R2 = 0.983 and R2 = 0.998. Then a graph to y_(C_Lt )/P and H_t/P a relation to calculate the critical depth were presented. The results showed that the calculated flow rate obtained using yc is consistent with laboratory values so that the relationship between them is R2 = 0.982.Labyrinth weirs are among the hydraulic structures that are constructed to regulate the water level and control the flow in canals, rivers and reservoirs of dams. This structure is designed to transmit large currents in low heads by increasing the effective length of the overflow crown. The crest axis of this type of weirs is indirect and in showing the horizontal surface, the overflow is composed of interconnected walls. Labyrinth weirs are repeated with triangular, trapezoidal, rectangular and arc geometries alternating in flow width. The main criterion in the labyrinth design of the weirs was to increase the flow transmission capacity on the overflow with a fixed canopy and for a certain height of the water level upstream of the weir. In this study, 145 laboratory data were used, including four types of arc labyrinth weirs with different arc radius and tangent lengths. The discharge efficiency was introduced as a dimensionless parameter  using dimensional analysis. The relationships between  and the critical depth of flow (y_C/P) were obtained using graphs for two different types, R2 = 0.983 and R2 = 0.998. Then a graph to y_(C_Lt )/P and H_t/P a relation to calculate the critical depth were presented. The results showed that the calculated flow rate obtained using yc is consistent with laboratory values so that the relationship between them is R2 = 0.982.

According to the research background, increasing the approaching speed has been reported as the most important factor in reducing the discharge capacity and also, the falling jet from the inlet and outlet keys has been reported as the main cause of foundation erosion in piano key weirs. Researchers believe that the geometric characteristics of piano key weirs have a significant effect on hydraulic behavior, The phenomenon of immersion on the crest, as well as the downstream hole scour of this type of weirs. so; The authors of the present study, in addition to investigating the hydraulic performance of Type A piano weir in standard and zigzag lateral crest conditions, considered it necessary to investigate the dimensions of the hole scour in the downstream of this weir. therefore; The effect of zigzagging of the lateral crest profile, which was modeled with the aim of improve transmission capacity and increase the immersion threshold in high water heads, was investigated on the geometric characteristics of the hole scour.

Experiments in a channel with a rectangular cross section, metal frame and glass wall, with a length of 10 meters, a width of 0.3 meters and a height of 0.5 meters and a slope of 0.0012 in the laboratory of hydraulic models Department of Water Science and engineering was done at Ferdowsi University of Mashhad. Based on the critical depth of the flow on the crest, Range of changes in input flow rate in the range Lit/s were set to achieve a minimum relative head of water on the weir crest with values greater than 0.1. According to this criterion, the relative water depth range of all experiments in the range ( ) was considered. Free flow conditions were adjusted by considering the range ( ) and up to the value by the sleeper valve downstream of the channel. Physical modeling of weirs was modeled with two forms of standard weir crest profile profile (type A) and zigzag crest with ratios P/Wu=1.33 and Wi/Wo=1.2 in 2.5 repetition with the standard sharp crest weir model with constant and equal height, it was used for experiments. In the second model, the form of the zigzag crest weir of the piano key was designed as a sinusoid with a height of 5 mm. During the lateral crest, 9 complete sinus zigzags were modeled. The sedimentary layer of the canal floor with two non-uniform and non-stick grains (1 and 3.2 mm), according to the control test results, was considered to be 0.4 m thick and 2 m long.

In the present study; By designing a sinusoidal zigzag in the lateral crest of a standard piano key (type A), it was shown that the maximum discharge coefficient occurs in a smaller ratio H/P but with a higher numerical value than the standard mode, so that the average discharge coefficient increases by 10% Improves discharge capacity efficiency. On the other hand; Due to the importance of scouring, changes in the geometric parameters of the hole and ridge sedimentary in the downstream erosion bed were measured and analyzed by dimensional analysis using Π-Buckingham method. The interaction of the output current jet in dealing with the sedimentary bed and intersecting with the overflow flow of the overflow inlet keys is the main cause of obstruction and the emergence of two rotating vortices in the lower hole erosion downstream of the piano key weir. The results showed that the maximum depth of the hole scour decreased by approximately 31% with the piano key weir of the linear sharp crest. Also, with the zigzagging of the lateral crest of the standard piano key weir (type A), despite an increase of 15% in the length of the hole erosion, the final scour depth decreased by 12% compared to the standard model to half the corresponding value in the linear sharp crest model. Furthermore; At maximum critical relative depth values of hc/P and hd/P water level compared to minimum values, the maximum scour depth increases by 73% and decreases by 90%, respectively. Also, it was observed that scour values occurred in fine-grained sediments 68% more than coarse-grained sediments. The process of sediment transfer in the hole and ridge scour increases with increasing The particle Froude number. In the particle Froude number range, Frd50 = 2.7-1.2 for the PKW, the average depth and length of the hole scour are estimated to decrease by 10% and increase by 22% respectively.

Finally, it should be noted that the proposed crest shape has reduced the maximum depth of the concrete slab in the economic design of downstream protection structures of Type-A PKW.

Lateral intakes are used to divert water from the main river. One of the crucial points in the design of intakes is to provide conditions that supply maximum intake with minimum sediment. In addition to dewatering from the outer bank of the river, it is recommended to build a sill at the inlet of the intake and also to use submerged vanes to remove the bed load to the intake. Submerged vanes are plate-shaped structures that are installed in the bed of rivers and canals at an angle to the flow. The primary function of submerged vanes is to create a secondary flow, so they play a significant role in controlling inlet sediment to the lateral intake. There have been many studies on the use of submerged vanes in front of lateral intakes. In all previous laboratory and numerical studies, the geometric parameters of the intake have been fixed, and laboratory-scale studies have been performed. In the present study, simulations have been performed in geometries close to natural conditions. Also, the intake is installed in different positions and angles, and by changing the width and height of the sill, the submerged vanes with two different arrangements are placed in front of the intake. The effect of vanes and change of intake parameters on the amount of sediment entering the intake and the anti-sediment coefficient have been investigated.

In the case of modeling in geometries close to natural conditions, the use of numerical models is necessary. The numerical model is used as a virtual laboratory, but in natural dimensions. Therefore, in the present study, numerical modeling has been used. The numerical model used (SSIIM2) solves the flow field by solving the Navier-Stokes three-dimensional equations using the finite volume method. To validate the numerical model, the junction of Kaskaskia River and Cooper River was simulated, and the bed changes predicted in the numerical model were compared with Rhoads (1996) field results. Comparing the results showed that the pattern of sedimentation and scouring in the numerical model is similar to field data. The numerical model has well predicted the location of scouring and sedimentation. In the present study, the intake has been installed in a 50-degree bend channel whose hydraulic dimensions and conditions are close to a part of Karkheh River. 27 numerical studies have been performed to investigate the effect of using submerged vanes on the value of the anti-sediment coefficient. Studies have been performed in three groups of 9. The first group had no submerged vanes (No vane), the second group had two rows of submerged vanes installed upstream of the intake (Layout1), and the third group had four rows of submerged vanes placed both upstream and in front of the intake (Layout2). In each category, the geometric parameters affecting the performance of the lateral intake that are dimensionless compared to the main channel parameters are the ratio of the intake width to the width of the main channel (Bi/Bm), the position of the catchment in the arc (θci/θc), intake angle (αi) And the ratio of sill height to flow depth (hs/hm). Each of these parameters is considered at three levels of change, as table1. In each of the 27 cases studied, the flow field is first dissolved, and after the convergence of the flow field, suspended sediment is injected from upstream on the rigid bed.

The radial velocity profile (ur) in the bend can be a measure of secondary flow strength. When the surface flow and the bed flow are opposite in both directions, the velocity profile is α-shaped, so a strong secondary flow is generated. If the velocity profile is β-shaped, the flow direction is in the same direction at the surface and the bed, and no secondary flow occurs. The closer the radial velocity profile is to the α-shape, the greater the secondary flow strength. The presence of the intake in the bend leads to the weakening of the secondary flow in the area affected by the intake and even the radial velocity is β-shaped, which is one of the factors in transferring sediment to the intake. The presence of submerged vanes have led to a change in the type of velocity profile from β to α. In general, the use of Layout1 has resulted in an average reduction of 15% in the amount of sediment entering the intake. In contrast, the use of submerged vanes upstream and in front of the intake, Layout2, has reduced the average amount of sediment by 46% to the intake. In studies (Barkdoll et al., 1999) that investigated the effect of using submerged vanes in controlling inlet sediment to the lateral intake on a straight channel with alluvial bed, three rows of submerged vanes were used upstream and in front of the intake. The results of theirs studies showed that for the discharge flow of 0.16 (Compared to discharge flow 0.25 of the present study), the submerged vanes led to a 35% reduction in the sediment ratio to the intake. For the flow ratio of 0.24, the submerged vanes caused a 50% reduction in the sediment ratio to the intake.

In the case where Bi/Bm>=35%, Layout1 did not affect the value of the anti-sediment coefficient. That is, if the intake’s width is large, it is necessary to install submerged vanes in front of the intake to be able to affect the amount of sediment entering the intake. In general, by changing the levels of the parameters θci/θc, αi, and hs/hm, Layout1 has led to an average increase of 59% in the anti-sediment coefficient. Layout2, on the other hand, has resulted in an average 148% increase in the anti-sediment coefficient compared to the case where no submerged vanes are used in the bend. Keywords: Lateral intake, bend, numerical simulation, SSIIM.

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 artificial laboratory channels 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 channels originated from 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 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.

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 28000 and 0.34, respectively. Therefore, the flow in the compound channel in the present study has turbulent and subcritical regime. 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 images, the floodplain and main channel bottoms were coated with black color in the measurement zone. Moreover, for detection of 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.

Chi-Squared test were used to determine the appropriate distribution to describe the longitudinal and transverse velocity and acceleration of individual particles (Fig. 5). Also, skewness and kurtosis of data series are employed to investigate the fitness of velocity and acceleration data to the normal distribution (Eqs. 2 and 3). The skewness values for the particle longitudinal and transverse velocities are always close to zero and their kurtosis values are close to 3, in the case of sediment release at 20 cm from the floodplain side wall. This indicates that the particle longitudinal and transverse velocities follow the normal distribution. However, kurtosis of longitudinal acceleration diverges from 3, and consequently, 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 transverse 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.

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 normal distribution. The histogram kurtosis of velocity and acceleration is enhanced approaching the interaction zone.