Computational Fluid Dynamics Service
ANSYS FLUENT software contains the broad physical modeling capabilities needed to model flow, turbulence, heat transfer, and reactions for industrial applications ranging from air flow over an aircraft wing to combustion in a furnace, from bubble columns to oil platforms, from blood flow to semiconductor manufacturing, and from clean room design to wastewater treatment plants. Special models that give the software the ability to model in-cylinder combustion, aeroacoustics, turbomachinery, and multiphase systems have served to broaden its reach.
Today, thousands of companies throughout the world benefit from the use of ANSYS FLUENT software as an integral part of their design and optimization phases of product development. Advanced solver technology provides fast, accurate CFD results, flexible moving and deforming meshes, and superior parallel scalability. User-defined functions allow the implementation of new user models and the extensive customization of existing ones. The interactive solver setup, solution and post-processing capabilities of ANSYS FLUENT make it easy to pause a calculation, examine results with integrated post-processing, change any setting, and then continue the calculation within a single application. Case and data files can be read into ANSYS CFD-Post for further analysis with advanced post-processing tools and side-by-side comparison of different cases.
Fluid Dynamics Simulation: Simulation Fluid Mechanism in Horizontal Separator
In this CFD analysis, the separator is modeled in CAD software and the grids are generated in GAMBIT using finite volume method. Fluent is used as solver and post processing software to solve the governing equation. The governing equations solved in this simulation are continuity equation, momentum equation and standard k-epsilon turbulence equation. VOF (Volume of Fluid) multiphase method is used to accommodate the nature of separation phenomena of three phase fluid inside the separator. The grid generated in this simulation is about 354000 cells to ensure the numerical validity of the simulation results.
Simulation is divided in three cases varied by the water cut of the entrance fluid (20%, 50% and 90%). The level of water is measured along the weir height to predict the maximum water level which still allow for the separator to operate properly and efficiently in separating these three phase mixture fluid into three single phase fluids of oil, water and gas. Simulation is conducted for 10 minutes of separator operating time with 1 second of time step.
Results and Conclusions
The grid generated in this simulation is about 354000 of tetrahedral cells as seen in Figure 1. The tetrahedral mesh is chosen because of their simplicity and flexibility to cover up the curve and sharp geometries. However, those benefit cost on the long iteration time. After processing the calculation for 52 hours of CPU time, the iteration is converged and reaches 10 minutes total time step.
Fig.1 The grid topologies of the horizontal separator model.
CASE I: Water Cut 20%
Figure 2 illustrates contours of volume fraction of oil after 600 seconds of time step. This figure shows the separation phenomena of oil, gas and water from the mixture inside the separator. This separation process occurs by the differences in the mixture componentâ€™s density. It can be seen that fluid with blue color (water) occupies the bottom of the separator, especially in water compartment (left side of the weir). The water density is the biggest among the other fluids (oil and gas) composing the mixture. The weir inside the separator is placed as a border between water and oil compartment. Oil is represented by green color; it is occupying the middle of the separator on the water surface. Oil has lower density compare to water, hence it will rise up to the water surface by the gravitation. At certain level of the weir the oil will over flowing passing the weir and flowing to the oil compartment (right side of the weir). The red and purple is coloring gas in the mixture. Gas is occupying the upper part of separator caused by its density which is lowest among the other phases.
Fig.2 Contours of Volume Fraction (oil) at 600 seconds for water cut 20%.
Fig.3 3D view of contours of volume fraction of oil at the weir plane inside the separator.
Fig.4 Oil level at weir plane coloring by contours of volume fraction.
Figure 3 is showing the oil level at the weir plane in water compartment side in 3D view, while Figure 4 illustrates the detail level of oil at this weir plane coloring by volume fraction. From Figure 5, the height of weir plate is 1.1 m and the interface of oil and water is laying at about 0.4 m. It means that the height of weir plate is very safe for separating oil from water at water cut 20%.
CASE II: Water Cut 50%
Figure 5 illustrates the separation phenomena of fluid mixture with water cut 50% into three different phases. The separation phenomenon is obtained after about 10 minutes of separator operating time. Water is represented by the blue color, oil by green to yellow color and gas by red and purple. It is shown for water cut 50%, more water occupies bottom of the separator in water compartment.
Figure 6 shows the water level on weir plane inside separator at water compartment side. It is shown that for water cut 50%, the interface of water and oil lays at about 0.77 of weir plate height. It means that even for water cut 50%, the separator is still operating properly in separating oil from water by the weir plate height of 1.1 m.
Fig.5 Contours of Volume Fraction (oil) at 600 seconds for water cut 50% .
Fig.6 Contours of Volume Fraction (oil) at plane weir level.
Fig.7 3D Pathline Colored by Volume Fraction (oil) at 600 seconds for water cut 50%.
Figure 7 shows the pathline of fluid flow in the separator picturing the sparation of oil, water and gas after 10 minutes operating time. Oil with its lower density than water occupies on the water surface and at certain level of the weir height, the oil is flowing on the top of weir height to the oil compartment.
CASE III: Water Cut 90%.
Figure 8 illustrates contour of volume fraction of oil inside the separator. This figure shows the separation phenomena of crude oil with water cut 90% in horizontal separator. In this figure blue is representing for water, green for oil and purple for gas. It is shown that for this case, the water is filling up the water compartment. A large amount of water are also occupies oil compartment. This condition illustrates that the separator is no longer effective in separating oil from water. It is caused by the large water content (90%) in the crude oil. In this case, the oil line must be bypassed to the water line.
Fig.8 Contours of Volume Fraction (oil) at 600 seconds for water cut 90%.
Fluid Dynamics Simulation: Fluid Mechanism in Centrifugal Pump
In this CFD analysis, the centrifugal pump is modeled in CAD software and the grids are generated in GAMBIT using finite volume method. Fluent is used as solver and post processing software to solve the governing equation. Fluent is used as solver and post processing software to solve the governing equation.
These following equations & laws are used in this simulation to solve the case:
- Conservation laws
- The Navier Stokes and Euler Equations
- Standard k-e Model
- Transport Equations for the Standard k-e Model
- Modeling the Turbulent Viscosity
- Model Constants
- Energy Loses in Fluid Flow
- Working Mechanism of a Centrifugal Pump
- Generation of Centrifugal Force
- Conversion of Kinetic Energy to Pressure Energy
- Pressure to Head Conversion formula
- Power and Efficiency
- Performance Curves
Results and Conclusions
There are three centrifugal pump models under investigation, so the results are also divided in three parts.
a. Centrifugal Pump Water liquid (water transfer pump) Model
- pump suction pressure = -0.12 = 0.91 kg/cm2 Abs
- RPM = 3540 rpm
- liquid density = 1015 kg/m3
- liquid viscosity = 1 cP
- BHP at 60% eff est = 20 kw
Fig.1 Contours of velocity magnitude (m/s).
Figure 1 shows the velocity of the curve casing area after leaving the house of impeller is higher than the zone of flow stagnation area on the casing or volute. And also the magnitude of velocity in the lower blade surface shows the highest magnitude. So that in this region could be said for having the suction, because pressure gradient is lower than the surrounding areas.
Contours of static pressure magnitude (Pascal).
Figure 2 illustrates the flow stagnation zone with the highest values are in the red area. This indicates that the flow due to the collision of the wall so the momentum of the flow decreases. On the side of the blade, there are two areas, suction area which is located at the lower of each blade and the pressure area at the top of the blade. In the figure above that the suction area on the blade is exactly at the surface of blade that colored green. And the pressure is on the top of the blade surface, the blade shown in red indicate areas with high pressure.
Figure 3 shows that in the discharge partially reverse the flow, lines of
flow direction appear partly in and partly out again. This shows the effect
of adverse pressure gradient. Adverse pressure gradient resulted flow in unfavourable
condition. So the more distance the flow will separate because the flow kinetic
energy is not able to resist the effects of the viscous stress and negative
pressure. From the figure above the point of separation seen in front of outlet.
Fig.3 Vector of velocity magnitude (m/s).
Fig.4 Pathlines of static pressure magnitude (pascal).
Figure 4 shows that the flow stagnation zone with the highest value is in the red area. This indicates that the flow due to the collision of the wall so the momentum of the flow decreases. On the side of the blade, there are two areas, suction area which is located at the lower of each blade and the pressure at the top of the blade. In the figure above that the suction area on the blade is exactly at the surface of blade that colored green. And the pressure is on the surface of the blade, the blade shown in red indicate areas with high pressure. The static pressure contours in the blade front surface (working surface) are shown in Fig. 4. It is found that the pressure increases from the inlet to outlet because of the work of the impeller blade. The total pressure is similar. But the static pressure on the back blades surface shown in Fig. 2 is different.
Fig.5 Pathlines colored by velocity magnitude (m/s).
Figure 5 shows that the velocity of the curve casing area after leaving the house of impeller is higher than the zone of flow stagnation area on the casing or volute. And also the magnitude of velocity in the lower blade surface shows the highest magnitude. So that in this region could be said for having the suction because pressure gradient is lower than the surrounding areas.
b. Centrifugal Pump Water liquid (injection pump) Model
- Pump suction pressure= -0.12 = 0.91 kg/cm2abs
- Rpm = 44.250 rpm
- Liquid density = 1015 kg/m3
- Liquid viscosity = 1 cp
- Bhp at 60% eff est = 275 kw
Fig.6 Contours of static pressure magnitude (Pascal).
Fig.7 Pathlines of static pressure magnitude (pascal).
Figure 6 shows a three dimensional view of the pressure over the shroud, blades and part of the volute for the nominal flow point. It can be seen that pressure in the hub side is higher than in the shroud, due to transaction of the flow from the axial to the radial direction. Figure 7 is the pathline of static pressure that show the flow impact over the tongue varies from the center – where the flow comes directly from the impeller to the sides.
Fig.8 Velocity vector colored by velocity magnitude (m/s).
Figure 8 show the velocity vectors at the blade trailing edge are plotted in figure 8, from hub to shroud for the three flow rates. The lower flow rate shows a tendency to separation in the suction side while, for the higher flow rate, this tendency happens in the pressure side. This effect is more evident in the hub section than in the shroud one, were relative velocities are less influenced by the flow rate. Fluid flow phenomena be shown in figure 9 pathline of fluid flows.
Fig.9 Pathlines colored by velocity magnitude (m/s).
c. Centrifugal Pump Oil Model
- Pump suction pressure = -0.02 = 1.02 kg/cm2abs
- Rpm = 3000 rpm
- Liquid density = 800 kg/m3
- Liquid viscosity = 4 cp
- Bhp at 60% eff est = 16 kw
Fig.10 Contours of absolute pressure magnitude (Pascal).
Figure 10 show the pressure distribution over the suction and pressure side of the blades is clearly appreciated. Near the shroud, the channel between blades is shorter and less curved, increasing the pressure gradient. It is visible the circumferential change because of tongue interaction: pressure at the channel outlet decreases in the anti clock wise sense, suddenly raising when the channel confronts the tongue. These differences are larger for higher and lower flow rates.
Fig.11 Contours of velocity magnitude (m/s).
Fig.12 Velocity vector of velocity magnitude (m/s).
Fig.13 Pathlines of static pressure magnitude (pascal).
Fig.14 Pathlines of velocity magnitude (m/s).