The Study and Analysis of the Effect of the T-shaped Groin Length on the Riprap Stability around the Rotatory Groin of the Rivers

1Assistant Professor, Department of water sciences & engineering, college of agriculture; Jahrom University, Jahrom, I.R. Iran. Postal code: 74137-66171 2Assistant Professor, Department of Water Science and Engineering, Ahwaz Branch, Islamic Azad University (IAU), Ahwaz, Iran. 3 Assistant Professor, Civil Department, Islamic azad university, Sepidan branch, Iran email : roozbeh1381@yahoo.com (*: Corresponding Author, Email: mrrafiee73@gmail.com) Abstract


Introduction
Using impermeable groin is a common methodsfor river bank erosion protectionleading to bank erosion control along with recovery and restoration of the valuable lands in the rivers' margin, in case of accurate design and implementation. Groin is a relatively economical and appropriate structure which have been used from the past for different organizational purposes of the river. Extensive application of groins(sometimes accompanied by their deterioration due to the lack of enough knowledge about design issues), has attracted many researchers of the hydraulic science. Although, the groins structures are formed with the aim of sedimentation and avoidance of the bank and margin erosions of rivers and stabilization of rivers situation, they themselves are affected by the erosion phenomena derived from the flow centralization especially in the nose section. In the nose section, local increase of the flow velocity derived from narrowing of the section and the formation of downward rotatory flow, leads to the formation of the scour hole of which advancement endangers the structure strength. According to Lags et al. (2001), the stile breakdown is divided into three groups which are the stile elements erosion, sub-surface erosion, and mass erosion. Erosion of the stile elements is affected by the size of the stile rocks in comparison with the hydrodynamic and disturbance forces, however, the stile slope, stroke, trituration, freezing or wave can also lead to this kind of destruction. Sub-surface erosion happens when the bed materials that are smaller than the stile elements leave the stile void and lead to the stile subsidence. Using filter reduces this kind of erosion to a large extent. Mass erosion happens when a big section of the stile elements or the bed materials collapse as a result of the slippery gravity forces. Based on the experimental results of Chiew (1995), three different mechanisms contribute to the stile instability around the truss including the break derived from cutting, sub-surface demolition and the edge demolition. The break derived from cutting is the stile elements mentionedby Lags et al. (2001). Edge demolition is the moving of the stile elements towards the formed scour pit in the bed sediments leading to the reduction of the conservative effect of the stile. According to Unger and Hager (2006), among the three mentioned mechanisms, only sub-surface slide and demolition in the stile around the groin were observed. Regarding that, in the present research, the flow is of a stormy typeand the drag force is calculated as: (1) Considering Reynolds analysis for the velocity parameter (u=û+ú) in the stormy flow, the velocity oscillation and average terms which are of the important components in the recognition of the nature of the sediments moving threshold can be reached.
The amount ofvelocity oscillation rate (ú) can be considered the same as the average velocity term (û). The drag coefficient and the cross section before the flow are two very important parameters in the calculation of the expulsion force (drag. The amount of velocity oscillations is dependent on the selection of these two parameters, however it is notable that flow velocity is very variable in vertical direction adjacent to the bed, effecting the drag coefficient value (Dwivedi, 2010). (Dwivedi, 2010). EsmaeelNejad (2003)-studied the hydraulic input flow to the basins of the deflective dam with a micro-model. In order to prevent the problem of sedimentation in the -basins, he introduced changing of the Chamran basin situation, reduction of the width, along with the increase of the -basin bed level, balance continuously as the most appropriate option. Niknam (2013) simulated the sedimentation process in the Dez dam tank using CCHE2D software and showed that the main coarse particles deposit at the upstream of the dam and fine-grained ones at its downstream. Main sediments due to the high depth of the tank and the low velocity of the flow deposit in the upstream. Marelius and Sinha (1998) showed that at the angle of 40 degrees for a single sheet, the amount of scour around the surface increases but produces the most vortex flow. Abdelazim et al. (2010) estimated lifetime of Aswan dam on the Nile Riverusing the CCHE2D two-dimensional model. In their research, they performed a simulation for 150 km of the tank length (from 350 km to 500 km) and estimated the lifetime ofof the tank as 254 years. The groin length is selected based on the amount of the width reduction of the main channel of the rivernot exceeding 30 %. The distance between the groinsis often selected as a coefficient of their lengths, for example Holland Delft hydraulic laboratory suggested the distance of 1 to 5 times of the groin length. Richardson et al. (1975) has reported the ratio as at least 3 to 4 and utmost 10 to 12. Charlton (1982) has suggested the ratio for relatively direct ranges as utmost 4 or 4.5 which can increase for internal walls of the turn and decrease for the external walls. Kynvry and Mvvrash (1984) recommended the ratio for protection of walls in the range of 1 to 5 times of the river's length. Peterson (1986) has suggested the ratio for protection of the walls in the ranges of 2 to 6, for shipping in the ranges of 1.5 to 2. The T-shaped groin length on the riprap stability around the groin in the shipping rivers using CCHE2D model was evaluated and compared with the measured amounts in the physical model.
Mohammad Ismail Nejad of the year (1381), by creating a micro model hydraulic inlet flow to the basin diversion dam HAMIDIEH studied. His sedimentation in the basin CHAMRAN and AZADEGAN to fix the problem, change the position CHAMRAN the intake, reducing the width and increasing the intake of AZADEGAN on both floor level, has introduced the most suitable option. HadiAliNaqi Zadeh Behbahani in the Year (1387), using physical models to study the hydraulic conditions in the intakes of the angle 90 and 75 degrees. The results show that the size of the deflection angle of deviation flow rate will have a great impact on the so that the angle of inclination of 90 ° to 75 ° in the same condition, is further deviation than rate. The location intake (in terms of angle deviation), a great impact on sediment deviation ratio so that Angle of 75 ° to 90 ° in the same hydraulic condition, the ratio of deposits constituted less deviation. Taebi et al (1388), the numerical simulation of flow in a 90 degree arc analyzed using the model CCHE2D respectively. The results showed that the high speed range at the beginning bent to an angle of 50 ° arc in the inner wall of the flume and then 50 degrees after to the middle and at the end of the flume is deflected toward the outer wall. Azarang (1388), in their study simulate the hydraulic and sedimentary Karun River in Ahvaz range Farsiat to help one dimensional CCHE2D software is studied. The results of the model CCHE2D, the sedimentation in the range of Ahvaz -Farsiat was estimated at about two and a half million tons per year. Roughanian in year (1389), Karun river flow hydraulic properties may be evaluated using the model CCHE2D and found that in the lateral arch velocity and shear stress than other parts (inner) more. Kamanbedast  , showed that at an angle of 40 degrees for a single page, although the amount of scour around page will increase, but most rotational flow is generated.
Monadyzadeh in year (1392) Laboratory results dike impact on the flow pattern on the verge of The Agribusiness Dehkhoda the intake using micro model examined. For this purpose micro model of the study area was constructed in the laboratory and testing was done on it. The results show that if dike at distance width of river at intake mouth to distance of 82 meters from the center of the intake to be built in its upstream, dike will have the greatest impact on enters rate the intake and increased 35 percent by volume of water entering the intake. Duan (1998), the CCHE2D model that was called CCHE2D upgraded model and in which characteristics of secondary flows were formulated using CCHE2D model for transformation of two-dimensional to threedimensional flow used. Barakdl and et al (1999) examined ways of increasing the efficiency of submerged vanes. Tests conducted by he showed that the intensity of input seabed sediments into the basin after installing of submerged vanes, when is low which flow rate ratio in the width units basin to flow rate the main channel width, is less than about 0.2.

Laboratory equipment
To calibrate and compare the simulations, a laboratory flume made in the Ahvaz Islamic Azad University was used to examine the riprap stability around the T-shaped groin factors in the rivers' arc with the varying dischargesof 1.85, 3.75, 5.08, 6.15 liter per second on the 90 degrees rivers' arc. - Figure 1 shows the entrance and outlet channels of the flume made in a rectangular section with plexiglass sheet walls.The length of the direct entrance channels at the beginning of the flume and the outlet at the end of the flume are 4.5 and 2.5 meters, respectively. The width and height of the flume are also 0.5 and 0.6 meters, respectively and the flume body is constructed with 0.7 meters height above the ground. The floor of the direct entrance passage (to create monotonous and developed flow) has been made of a metal sheet with a thickness of 10 mm and a length of 4.5 which reduces the effect of the roughness of the wall and also the hydraulic phenomenon in the tank is observable. The outlet passage is also direct, its floor is made of a metal sheet with a 3 mm thickness and its wall is of plexiglass with 10 mm thickness and 2.5meter length. The flume floor is made in the form of a fixed bed and without longitutinalslope. After relaxation of the considered input flow, the flow should enter an arc with 90 degrees angle with external radius of 2 meters. It is made of a plexiglass sheet with a 10 mm grade so that they can be implemented easily in the arc and for the flume hermetic sealing, the aquarium glue was used.  Small concrete pieces were put together in a simple and T-shaped form and used as groins (Fig.2) The main channel of the flume consisted of an initial direct channel with 4.5meter length before the arc and an arc with an external radius of 2meters and a second direct length of 2.5meters after the arc and a width of 0.5 meter. Water enters a P.V.C tube with 3inches diameter using an EPS pump and enters the tank from under the flume initial tank in the form of a spring. Flow discharge to the flume was fixed by a gate valve installed at the entrance of the main channel.The inflow discharge was assessed manualy by measuring flow depth overa triangular overflow with a 90degrees openingheadinstalled at the canal inlet-After tankage from the basin at different deviation angels, amount of the output water passes the flume over the terminal triangular overflow in order to be measured again which by subtraction of these two rates of flow, the input discharge in different manners of experiments will be obtained. Finally, the outlet flow reaches the terminal tank and will be returned to the system by the installed P.V.C tube under the tank to the underground resource for the laboratory water supply and also the entered discharge to the basin will enter the water supply underground resource again using an appropriate pump. The research data can be classified into two series as: 1) The initial topographic data of the flume and basin 2) The flow data The initial topographic data are used to define and interpolate the required Mesh network datanecessery for the simulation. The process of sediment transport modeling includes flow simulation in which the process of transmission and distribution of the sediments in the basin are simulated. Actually, the flow and sediment simulation are performed concurrently and in doubles.  (Table 1). Analysis of the effective factors shows that the stability of the stile and eventually the groin can be dependent on the following factors. In which the equals to 90 degrees.
DStile geometry including relative diameter and density (stability number) The index (Nc) was used to examine the ripraps stability.
Results and discussion 1. Examination of velocity, shear stress, and the ripraps stability forgroin length of 20 and wing length of 12 % of the useful flume width.

Examination of velocity, shear stress, and ripraps stability forgroin length of 20 and the wing length of 24%of the useful flume width
Regarding the Table (3) and diagram of the figures (6) and (7), it can be observed that with increasing Froude number from 0.032 to 0.171, inceases the average flow velocity from 0.031 to 0.167 ms -1 and the flow velocity near the T-shaped groin nose has increased from 0.036 to 0.169 ms -1 . A comparison between velocity values shows that velocity values in the groin nose are higher that those of average as1.8%. Also, by the increase of the Froude number from 0.032 to 0.171, the average shear stress has increased from 0.01 to 0.278 Nm -2 and the shear stress near the T-shaped groin nose has increased from 0.012 to 0.266 Nm -2 .