---

In order to facilitate the release of floods from the dams and to prevent their damage or collapse, a structure called a spillway is used. Due to the natural and variable flow of the input to the reservoirs of the dams, there are times when the river inflow exceeds the consumption amount in the downstream agricultural lands. In these cases, excess water is discharged over the crest of the weir and flows towards the spillway, which causes high velocities. This high velocity creates low pressure areas on the spillway concrete surface, which can cause major damage to the spillway or even endanger the integrity of the dam structure. Therefore, the dam spillway must safely dissipate the kinetic energy. One of the types of weirs is the stepped spillway to facilitate the passage of the flow over the dams. One of the most obvious practical features of stepped spillways compared to other spillways is the considerable energy dissipation along the spillway. Care should be taken in designing and selecting the type of spillway to prevent potential erosion and reduce kinetic energy as the water flow passes over the spillway. One possible solution is to use a stepped spillway instead of a smooth spillway. In this study, a numeral model of a stepped spillway with different steps and slopes is used. For this purpose, ANSYS software is used for modeling free surface with application of k-ε turbulence model. In the present study, numerical simulation using the Volume of Fluid (VOF) model was used to investigate the mixing phenomenon of two phases of air and water of the free surface flow. The flow field was continued until the residuals reached 10^-7. Compared to simpler models such as Mixture, which operates solely on the basis of averaging the properties of two phases, the VOF model, is separating the phases and considering the effects of the interface. The VOF model, is capable of more accurate simulation of phenomena such as fluid mixing, turbulent flows, and heat transfer in multiphase flows. A number of hydraulic specifications which are considered in designing the stepped spillways are the pressure on the surface of the steps, velocity distribution and energy dissipation. The results from the numerical models were compared with experimental studies. They showed acceptable agreement with physical simulations. Results show that discharge and spillway slope increment reduces the amount of energy loss. In the spillway with 5 steps, for a discharge of 0.063 m³/s, the amount of energy dissipation at a slope of 26.6 degrees changes from 85 to 82% at a slope of 45 degrees, which shows a decrease of 3%. With the increase in discharge, the flow depth increases and reduces the effect of the roughness of the steps on the upper layers of the flow. Increasing the height of the steps increases the rate of energy dissipation and also increases the occurrence of negative pressures in stepped spillway. In this case, the contact surface between the main flow and the eddy currents increases. With the increase in the height of the steps, the dimensions of the rotating vortices also increase and cause a larger radius of rotation on the steps. The presence of these large rotating vortices separates the flow from the bottom of the steps and reduces the pressure on the surfaces. The number and dimensions of steps can alter the energy dissipation rate. Increase in the number of steps in a spillway with constant height, reduces the energy loss as the result of steps dimensions being shrunk

---

A passive scalar is a property that is affected by the flow field without affecting it. In this paper, first, the governing equations on the turbulent flows are solved and the property of a passive scalar in two dimensions is numerically studied. Having the values of the velocity components, the governing equation on transport of a passive scalar is solved. To compute the turbulent velocity field, the Large Eddy Simulation (LES) method using Smagorinsky subgrid scale is invoked. The flow in a cavity has been the basis to validate the accuracy of the generated computer code. To ensure the compatibility between the flow and the transport of passive scalar fields a similar LES approach is used. As a three-dimensional numerical solution for a turbulent flow fields needs a massive computational time and efforts, therefore a two-dimensional simulation used for a proper saving. Instead, to validate the numerical results, the range of the Reynolds number of the flow is kept within the range of the two-dimensional measurements. Comparison of the numerical results and the experimental measurements in two-dimension reveals the high accuracy of the results and compatibility between the flow and passive scalar fields. Ability of developed scheme to accurately handle transport of a passive scalar is promising to extend LES method into the transport of more species and hence to simulate reacting flows.

---

---

The flow at a channel bifurcation is turbulent, highly three-dimensional (3D) and has many complex features. There is transverse motion accompanying the main flow and an extensive separation zone that develops in the branch channel. There are two complex flow regions along the intake channel: one is the separation zone and the other is the region in which helical motion of water particles forms. This separation occurs because the flow entering the branch channel has considerable momentum in the direction of the main channel flow. This zone causes hydraulic and sedimentation problems that must be known before designing the system. This necessitates a deeper insight into the flow patterns and shear stress distributions near the solid boundaries. In this research, 3D flow patterns at lateral diversion were investigated experimentally and numerically. The experimental investigation was carried out at a T-junction, formed by two channels with rectangular cross-sections. The width of lateral intake to the main channel was 0.4. 3D velocity measurements were obtained using Acoustic Doppler Velocimeter at junction region for 11%, 16% and 21% discharge ratios. Fluent mathematical model was then used to investigate the dividing open-channel flow characteristics. Turbulence was modeled by Two Equation (k-ε, k-ω) and Reynolds Stress (RSM) turbulence models. The predicted flow characteristics were validated using experimental data and the proper model was selected for hydrodynamic and parametric studies. Within the main channel, good agreement was obtained between all models prediction and the experimental measurements, but within the lateral channel, the RSM predictions were in better agreement with the measured data, and k-ω predictions was better than those of k-ε. The comparison of experimental and numerical streamlines at different elevations showed that the selected model is capable to simulate the most important features of flow at diversions. The study of the velocity contours at different elevations showed that the velocity magnitude decreases at main channel, just downstream corner of lateral intake at the near bed levels and this causes the sedimentation in movable beds. The results showed that the width of separation zone at lateral intake will decrease and the distance of dividing stream surface from left bank of the main channel will increase by increasing of the discharge ratio. Investigation of the flow pattern at the entrance of the lateral intake showed that the secondary flow will form at this section. The dimension of the secondary flow at near bed elevation will increase by increasing of the discharge ratio and this causes entering of more bed load into the lateral channel.

---

In this article, a special duct is introduced in which, inlet water jet initiates to oscillate after a short time and it causes the velocity and pressure to oscillate regularly. Considering that there is a linear relationship between the inlet jet velocity and its oscillations frequency, the flow rate can be calculated by measuring the pressure frequency. In order to study the flow field inside the current geometry of fluidic oscillator and also to find the optimum location for sensor to detect the pressure oscillation, the unsteady turbulent Navier-Stokes equations are solved by ANSYS CFX software. Having studied the grid independency, capability of K-ε and SST turbulence models for numerical simulation of unsteady flow inside the fluidic oscillator is considered. Then, according to the peak to average ratio (PAR) criterion, the qualities of pressure signals are compared at some points, to distinguish an optimum pressure sensor position. Afterwards, a prototype of fluidic oscillator flow meter is manufactured for the first time in Iran. Using this prototype and inserting the pressure and Piezoelectric sensor at the optimum point, the numerical simulation results are validated by the experimental data. Comparison between the numerical and experimental results shows that the SST model is more suitable for this flow simulation. Finally, by performing experiments in different flows, acquiring and processing pressure signals, the flow meter characteristic diagram (inlet jet oscillations frequency- inlet jet velocity) are extracted.

---

This paper presents a comparative study of turbulence models performance in prediction the oscillating characteristics of naturally excited jet flows. The unsteady averaged Navier-Stoks equations for turbulent incompressible flow and five variant turbulence closures are used in this study. A large family of turbulence models exists in the literature which is far too extensive to be reviewed here. The models are ranged from simple algebraic expressions for the eddy viscosity to more elaborated formulations which introduce a separate transport equation for each component of the Reynolds stresses. The software, FLUENT 6.3.26, was employed for solving the governing equations. Computational results compared with reported experimental data. The standard k-ε and SST k-ω models clearly showed better results than the others. The accuracy of standard k-ε model decreases with decreasing the nozzle inlet velocity and it failed to predict the minimum excitation velocity (minimum excitation kinetic energy) in the self excited fluidic nozzle.

---

In a vertical wind tunnel, in order to prevent persons or an experimental model from falling, a protective screen should be installed at the end of the nozzle section. Since the air has the maximum velocity at this section, the pressure drop due to the protective screen will be significantly high. On the other hand, as the protective screen is alternatively exposed to dynamic forces due to free fall of the floating persons or the model, the screen wires will experience fatigue. To prevent this, multi strand cables should be used in the manufacturing of these protective screens. In this research work, using the momentum difference method, drag coefficient for the multi strand cables and circular rods has been measured and compared. For this purpose, a hot wire anemometer (HWA) with a one-dimensional probe has been used. Results show that at Reynolds number in proximity of 2 × 103, the drag coefficient for the multi strand cable exceeds that of a circular rod by 16% and that this amount decreases with further increase in Reynolds number. The trend is such that for Reynolds number of 104, the drag coefficients of the multi strand cable and circular rod are almost equal.

---

This paper presents pore scale simulation of turbulent combustion of air/methane mixture in porous media to investigate the effects of multidimensionality and turbulence on the flame within pore scale. A porous medium consisting of a staggered arrangement of square cylinders considered here. Results of turbulent kinetic energy, temperature, flame thickness, flame structure and flame speed are presented and compared at different equivalence ratios. The turbulent kinetic energy increases along the burner because of turbulence created by the solid matrix with a sudden jump at the flame front due to increase of the velocity as a result of thermal expansion. Also, it is shown that at higher equivalence ratios, the effect of turbulence within porous burner is highly significant phenomenon. Due to higher turbulence effects in higher equivalence ratios, the flame thickness increases by increasing the equivalence ratio which is in opposite of the trend observed in laminar flow simulation. Also, it is shown that the dimensionless flame speed and excess temperature is higher at lower equivalence ratios due to lower heat loss to the cold upstream environment of burner. Two dimensional structure of flame in the pores of porous medium is shown in the present study via isotherm lines.

---

In the present study, the effect of intra-pore turbulence within porous burnershas been investigated on combustion of methane/air mixture in such burners. A model is adapted to the porous structure to models turbulence flow. The GRI 3.0 chemical reaction mechanism is utilized for the combustion of methane/air mixture and radiative part of the solid phase energy equation is obtained using the discrete ordinate method. The numerical results show that the gas temperature obtained from turbulence model stays below the corresponding laminar model temperature all over the combustion region, and the flame thickness becomes wider in turbulence model. Although the CO emission are insensitive to laminar or turbulence model, the burning speed and NO emission predictions are found to be significantly improved when the effects of turbulence are taken into account.

---

Abstract: The two main characteristics of a porous bed are the bed material diameter – representing its coarseness - and the porosity which represents its permeability. In this paper the effect of bed permeability on the average structure of flow turbulence is investigated. Flow is modeled using an innovativecontinuum approach based on Volume Averaged Navier-Stokes Equations in several different channel bed porosities. Results of four different simulations with various porosities are presented. Bed permeability can be represented by the permeability Reynolds number, Rek, which is the ratio of effective diameter of porosity and the length scale of eddies near the bed. The Reynolds permeability number (Rek) is the best expression for the bed permeability quantity. In small Rekthe bed acts as a solid/rigid boundary and in large Rek, the bed will behave as a high permeable boundary with negligible viscosity effects. Under these conditions, the turbulence eddies along the flow are rarely observable. The reasons can be due to: 1) the mechanism of free turbulence transfer through permeable layer and/or 2) a considerable decrease in the average shear stress due to no wall-blocking and low-viscosity effects. The dominant characteristic of turbulence near a high permeable bed is relatively large eddy structures, probably originating from so-called Kelvin-Helmholtz instabilities. Suchflow pattern with large vortexes leads to high momentum exchange between the free channel flow and the two-phase flow in porous media below permeable bed. This process also increases the friction between flow and the substrate and therefore will result in moving the Logarithmic-low region slightly downward. In addition, it is observed that the log-low cur near a high permeable bed is 4% steeper than in solid and rough bed.

---

The impact of the different integral scales of two isotropic turbulent fields on the dynamics of a shear-free turbulent mixing layer is investigated. To this end, two-dimensional incompressible Navier-Stokes equation is numerically solved using pseudo-spectral method. Governing equations are considered in the vorticity-stream function formulation to guarantee the divergence freeness of the velocity field. Dynamics of the turbulent interaction is examined through relevant statistical parameters such as skewness and kurtosis of the velocity components and their spatial derivatives. Moreover, the efficiency of mixing is investigated by considering the length and curvature of the mixing layer. It has been observed that increasing the difference between the initial integral length scales of two isotropic turbulent fields increases the mixing and anisotropic level of interaction.

---

Investigations of the phenomena associate with the Fluid-Structure interactionin transonic turbomachines due to the presence of unstable flow behaviors have double significance. Severe restrictions of the experimental methods, has developed researchers approach in this field to Numerical methods. Nevertheless, using simple two-dimensional model to investigate the phenomenon of quality is inevitable because of high computational cost of numerical methods in aerodynamic and aeroelastic simulation of full model of turbomachines. In this paper transonic flow in fixed fan cascade and fan cascade with central blade vibration in Forced harmonic pattern is simulated and variations of turbulence characteristic patterns are studied. In order to prevent divergence of the solution and achieve more accurate results, the step by step algorithm is developed. On the other hand, spring methodology with linear torsional springs is used for movement of dynamic grid around the oscillating blade. Mesh quality is assessed by examining maximum Mach number and y+ variation. Compare the results with the available experimental data indicated a significant difference in the position of the vortices are detached and re-attached. This difference proves need to use a turbulence model is more accurate in terms of the wide separation. In this paper, effect of blade geometry, flow separation and central blade oscillation on flow pattern and turbulence characteristics of transonic flow have been investigated. Obtained results explain the effect of mentioned parameters on the turbulence kinetic energy and dissipation frequency.

---

Spur dikes are the training structures that may be used for river bank erosion protections. These structures may increase the navigation depth. The spur dike creates stable pool for aquatic habitat. Most of the previous researches focused on scour hole dimensions and the flowfield around the emerged spur dikes. As the spur dikes may be submerged during the floods, in this paper, the flowfield around a submerged T-shape spur dike was investigated using Vecterino apparatus. The experiment was conducted in a channel with 8m length, 60cm width and 80 cm depth at TarbiatModares University (Tehran- Iran). The parameters that were investigated in this paper includes, mean velocities, turbulence parameters and streamlines. The results of this study indicate that the high velocity region of the flow elongates to the far distance of downstream in the upper layers. There is a recirculating flow downstream of spur dike. Due to the effect of the spur dike overflow, the downstream recirculating flow, rotates in the opposite direction as compare to the emerged spur dike. The recirculating flow elongates to the far distance of downstream as compare to the submerged trapezoidal spur dike. Turbulence parameters analyzed in this research include normal Reynolds stress, bed shear stresses, the probability of each process and triple correlations. Maximum normal Reynolds stresses were observed at the upstream tip of the spur dike. In addition, the maximum shear stress was observed at the same region. The high stress region around the spur dikes showed a shear layer region around the spur dikes. Analysis of the probability of the turbulent bursting events in streamwise direction shows that ejection and sweep events are the most probable events in the upstream section of the spur dikes. In the streamwise direction, the interaction events are the most probable process near the upstream tip of the spur dike, while ejection and sweep events in the widthwise direction are the most probable events near the upstream tip of the spur dike. In the downstream recirculation zone of the spur dike, the probability of the events approximately are the same in the streamwise direction and interaction events are the most probable events in widthwise directions. Triple correlations presented useful information about the turbulent bursting process. The triple correlation analysis in widthwise direction showed that the ejection is the strongest event in shear layer region. Triple correlation analysis in the streamwise direction, showed that ejection was the strongest event in the upstream of the spur dike and in the region between near channel wall and spur dike wing. This causes sediment transport in streamwise direction as suspended load. The interaction events are the strongest events in the downstream recirculation zone, hence the sediments were deposited in this region.

---

Non-Newtonian fluid flows experience turbulent regime in some industrial applications. Several approaches have been proposed for numerical simulation of turbulent flows that each one has specific features. RANS turbulence models have reasonable computational costs, while include several sources of uncertainties affecting simulation results. In addition, developed RANS models for non-Newtonian fluids are modified versions of available models for Newtonian fluids, therefore, they cannot provide reliable estimation for viscoplastic stress term. On the contrary, DNS delivers accurate results but with high computational costs. Consequently, use of DNS data for estimation of uncertainty in RANS models can provide better decision making for engineers based on RANS results. In the present study, a turbulence model based on for power-law non-Newtonian fluid is developed and employed for simulation of flow in a pipe. Then, an efficient method is proposed for quantification of available model-form uncertainty. Moreover, it is assumed that uncertainties originating from various sources are combined together in calculation of Reynolds stress as well as viscoplastic stress. Deviation of the stresses, computed using RANS turbulence model, from DNS data are modeled through Gaussian Random Field. Thereafter, Karhunen-Loeve expansion is employed for uncertainty propagation in simulation process. Finally, the effects of these uncertainties on RANS results are shown in velocity field demonstrating the fact that the presented approach is accurate enough for statistical modeling of model-form uncertainty in RANS turbulence models.

---

In this study, a three-dimensional fluid field of an axial flow type micro hydro named Agnew has been investigated. The turbine installed at the Hydrulic Machines Laboratory (HML) of Iranian Research Organization for Science and Technology has been designed to generate 1 kw output power.All numerical simulations were performed using ANSYS CFX, a Computational Fluid Dynamic code, to investigate the performance parameters, such as efficiency and power, and results are validated against experimental data. Four different grid sizes are studied in accordance with the Grid Convergence Index (GCI) to investigate mesh independency of the solution. Results of several turbulence models were also examined to find out the Shear Stress Transport (SST) model in order to take into account the turbulence in the flow. Several turbulence models were also examined together with wall function in order to take into account the turbulence in the flow. A mixing plane interface plane was used to pass the disturbance of rotary domain to stationary domain. The obtained results show that a high resolution advection scheme, mixing plane to model the rotor-stator interaction together with a turbulence intensity of I=6% at the inlet, best matches with the experimental results. The difference between the efficiencies computed from both numerical approaches and experimental values may be ascribed to a numerical error, a model error or a systematic error.

---

In order to assess the effect of turbulence models in prediction of flow structure with adverse pressure gradient, steady state Reynolds-averaged Navier-Stokes (RANS) equations in an annular axisymmetric diffuser are solved. After selection of the best turbulence model, an approach for the shape optimization of annular diffusers is presented. The goal in our optimization process is to maximize diffuser performance and, in this way, pressure recovery by optimizing the geometry. Our methodology is the optimization through wall contouring of a given two-dimensional diffuser length and area ratio. The developed algorithm uses the CFD software: Fluent for the hydrodynamic analysis and employs surrogate modeling and an expected improvement approach to optimization. The non-uniform rational basic splines (NURBS) are used to represent the shape of diffuser wall with two to ten design variables, respectively. In order to manage solution time, the Kriging surrogate model is employed to predict exact answers. The CFD software and the Kriging model have been combined for a fully automated operation using some special control commands on the Matlab platform. In order to seek a balance between local and global search, an adaptive sample criterion is employed. The optimal design exhibits a reasonable performance improvement compared with the reference design.

---

Experimental study on effect of single spur dike side slope on flow structure carried out. In the study, 3D flow dynamics around the single spur dike with side slope of 75 degree has been studied experimentally in straight fixed bed channel. In order to analysis of flow characteristics, turbulence parameters flow around the scour have been measured using Acoustic Doppler Velocimeter (ADV). In behind of the spur dike, due to reduction of flow velocity that leads to formation of stationary zone, flow velocity will be increased in central zones. Therefore, a part of the flow is conducted upward and another part will be directed downward, where the pressure is lower. Downward flow is the cause of horseshoe vortex formation. Two velocity intensifier zones is formed, one is located in the main core of flow which is caused by reduction of flow width and the other one is a high velocity zone that is related to local velocity intensifying in outer layer of shear zone in the downstream of the spur dike. This is stemmed from increase in velocity of flow caused by reduction in flow width and leads to scour initiation from this zone. The maximum amount of “-ρ(u^' v^' ) ̅” stress element in shearing layer direction is occurred. Regarding to negative amounts of “-ρ(v^' w^' ) ̅” and “-ρ(u^' w^' ) ̅” stress elements, accumulation of deposits is happened in circular zone behind the spur dikes. Increases in both the speed, a core area to increase the speed of the flow and reduce the width of the high-speed flow and other areas related to the intensification of local rapidly down the breakwater and in the area the outer layer the shear is formed. The maximum mean flow velocity at the bottom of the breakwater 55/1 is the mean flow velocity is approaching. In the upstream region of the breakwater a rotating flow in the leg is shaped breakwater. It has small dimensions and rotational flow near the breakwater body rotation power was concentrated in the can be high. Upstream of the nose, the lines of the horseshoe vortex, the rotation around the nose, the hands move downward and adjacent layers come to be intertwined with the shear. The average flow field, the maximum kinetic energy in the central part of the channel and away from the shear layer form is the maximum kinetic energy of turbulence along the shear layer will occur. Increase in kinetic energy along the shear layer plays an important role in the sediment bed holds it down, so that the kinetic energy due to the formation of vortex turbulence the total flow depth have been developed. In behind of the spur dike, due to reduction of flow velocity that leads to formation of stationary zone, flow velocity will be increased in central zones. Therefore, a part of the flow is conducted upward and another part will be directed downward, where the pressure is lower.

---

This article presents a numerical investigation of fluid flow in one of the centrifugal pumps of pump-Iran Corporation. A computational fluid dynamics (CFD) analysis is performed by using the CFX software for a wide range of volumetric flow rates for two different rotor speeds of 1450 rpm and 2900 rpm and the numerical results of water are validated against measured values of head and total efficiency with an overall acceptable agreement. The obtained results have been obtained for crude oil as diagrams of head and total efficiency as the function of volumetric flow rate and other variables and compared with results of water. Numerical results show that the absolute pressure on blade surfaces for crude oil is 705 kpa less than when using water. The absolute pressure difference between inlet and outlet of impeller and spiral volute for crude oil is comparatively less than those amounts in comparison with water. Also by increasing the angular velocity of rotor, it was observed that high levels of turbulence intensity are transmitted from outlet pipe bending to the impeller outlet at volumetric flow rate of 30 m3/h that causes the efficiency reduction and high levels of turbulence intensity for crude oil are less than those amounts in comparison with water within impeller area. Finally, to represent an impeller pump head curve for crude oil over the overall operation range of the pump, a second order polynomial equation was fit to numerical data.

---

Spillways have long been of practical importance to safety of dams, therefore these structures have to be built strong, reliable and highly efficient. Ski jump dissipator is one the flow energy dissipators which is applicable downstream of spillway chutes with velocity over 20 m/s. Flow over a flip-bucket is a two-phase and strongly turbulent flow. Turbulence modeling is one of the most limiting factors in accurate computer simulation of flows. By fixing the grid resolution and the discretization scheme, the difference of computation time is mainly attributed to the turbulence model. The choice of turbulence model depends on factors such as the physics encompassed in the flow, the level of accuracy required, the available computational resources, and the amount of time available for the simulation. It is a fact that no single turbulence model is universally accepted as being superior for all classes of problems.
The main purpose of the present study is numerical investigation of two-phase turbulent flow over a triangular flip-bucket to evaluate effects of different turbulence models in this type of flow. Hence, using FLUENT® software, two dimensional Reynolds averaged Navier-Stockes equations have been solved in unsteady state. Different turbulence models consist of k-ε, k-ω and RSM; have been used. To simulate two-phase flow, volume of fluid (VOF) method has been applied.
Standard k-ε and stress-omega RSM models with low-Reynolds number modifications have the best performance among the other turbulence models. In standard k-ε model when low-Reynolds number modification was activated, the effects of molecular viscosity were taken into account in near-wall regions. Therefore, in low-Reynolds number k-ε model, maximum dynamic pressure over the bucket was predicted more accurately in comparison with standard k-ε model. Regarding modification in strain-pressure terms in turbulence equations, effects of anisotropic Reynolds stress tensor were taken into account in stress-omega RSM model with low-Reynolds number modifications. Thus, compared to other turbulence models, numerical results of this model are in a better agreement with experimental results. Different k-ε models could not predict the jet trajectory after the bucket very well. Due to blending function in SST k-ω model, this turbulence model effectively blended the robust and accurate formulation of the k-ω model in near-wall regions with the free-stream independence of the k-ε model in the far field. In estimation of maximum dynamic pressure over the bucket, this model had a better performance than standard k-ω model and relatively similar results to k-ε model. In addition, SST k-ω model has shown the best prediction of the jet trajectory among other turbulence models. Eventually, with respect to computation cost and accuracy of results, SST k-ω turbulence model has been introduced as the most suitable turbulence model to predict the flow pattern of a triangular flip-bucket.

---

Many models have been applied to slug flow using laminar flow condition. The results obtained from these models are not consistent with the physical behavior of slug flow. Furthermore, discussion on the turbulent models is very rare or not related to the such flow regime. The slug regime is a complicated regime with shear flow and high strain in addition to some vorticity at some sections of the flow. In the present attempt at first stage, the turbulent models differences, the initial assumptions to drive, privilages and shorcomings have been considered with details. Then, its consistency with the physics of slug flow was analysed with high accuracy. In the second stage, simulations using different turbulent models were conducted. The obtained results were compared to each other and with the experimental results of other investigators. Finally, the most consistent model with the physics of the slug flow was selected. The turbulent model of RNG k-ε showed more reliable in compare to other turbulent models. Thus, it was selected and used to obtain slug flow behavior with higher accuracy. The parameters as pressure distribution during slugging, slug mixture velocity, slug initiation time and position from the duct inlet with RNG model conducted and presented with detailed explanations.