Flow field analysis to improve a liquid/liquid separator vessel design using CFD Rodrigo Peralta, ESSS Lucilla Almeida, ESSS Karolline Ropelato, ESSS Thiago Anzai, Petrobras Robson Pereira Alves, Petrobras João Cláudio Bastos, Petrobras Erick Quintella, Petrobras
24
Embed
Flow field analysis to improve a liquid/liquid separator … field analysis to improve a liquid/liquid separator vessel design using CFD Rodrigo Peralta, ESSS Lucilla Almeida, ESSS
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Flow field analysis to improve a liquid/liquid
separator vessel design using CFD
Rodrigo Peralta, ESSS
Lucilla Almeida, ESSS
Karolline Ropelato, ESSS
Thiago Anzai, Petrobras
Robson Pereira Alves, Petrobras
João Cláudio Bastos, Petrobras
Erick Quintella, Petrobras
Presentation topics
• Company overview;
• Problem description;
• Goals;
• Methodology;
• Results;
• Conclusion and next steps.
Company overview
Problem description
• Crude oil contains water and other contaminants which need to be removed for economical
transport and before further processing of the crude oil.
• Water in oil emulsion is a particularly problematic issue, since its formation makes it difficult
to separate with only the gravitational field, demanding, for instance, electrostatic devices.
Goals
• Simulate the multiphase fluid dynamic behavior of an oil/water
separator in different geometries configurations aiming further
construction;
– Several approaches were performed to improve separation process:
– inclusion of inlet device;
– The goal is to allow a smooth flow of the mixture inside the
equipment;
• Implementation of level control strategy with UDF;
• Implementation of simplified eletrocoalescence with UDF.
Methodology – Geometry
mixture inlet
oil outlet
brine outlet
• Initial geometry:
• Further configurations:
– Included an inlet device:
Simplifications: Electrodes and
perforated plates were not considered;
• Simulation result will predict their
location;
Methodology – Mesh
Mesh generated in ANSYS Meshing® software;
• Hybrid mesh: Tetra + Prism;
– Prism layer near walls to capture boundary layer effects and also in the expected oil/water interface
region;
• 1,46 millions of elements.
Prismatic elements
Finer mesh in regions of higher gradients
Software : ANSYS FLUENT 14.0®;
• Multiphase flow simulation
– Euler-Euler approach;
• Each phase has his own velocity flow field;
• Dispersed phase diameter:
– Study 1: 100 microns;
– Study 2: UDF to control this parameter inside equipment.
– Incompressible fluids;
– Continuous phase and dispersed phase interaction;
• Surface tension: 0,0377 [N/m];
• Drag: symmetric;
• Isothermal
• Turbulence: k-ε model;
• Transient.
Methodology – Mathematical model
Physical setup:
• Prescribed inlet velocity;
• Prescribed outlet pressure;
• Prescribed outlet velocity (controlled by an User Defined
Function);
• No slip walls;
Methodology – Additional variables for CFD analysis
• Flow uniformity coefficients analysis;
– Objective: Set the most appropriate position for internal devices, such as the
electrodes;
𝐶𝑉 =1
𝐴𝑡 𝐴𝑖
𝑢𝑥 − 𝑢𝑥𝑢 𝑥
2𝑁
𝑖=1
12
𝛼 = 𝑎𝑟𝑐𝑡𝑎𝑛𝑢𝜃
𝑢𝑥 𝑆 =
𝑢𝑥 𝛼 𝑑𝐴
𝑢𝑥 𝑑𝐴
0 m 4 m
Region of flow uniformity analysis:
Swirl number (S):
Coefficient of variation (Cv):
Measure the flow uniformity in tangential
direction.
• 100% uniform flow: S = 0;
Measure the flow uniformity in the flow
direction.
• 100% uniform flow: Cv = 0;
Results – without level control
Without a modified inlet device,
separation efficiency was lower;
• Small fraction of brine in the bottom of
vessel;
• Small fraction of oil in the top of vessel;
Including an inlet device:
• Improved oil/brine separation efficiency;
• Brine level still less than expected;
– Loss of oil by bottom outlet.
Low separation efficiency due to low
droplet size (without eletrocoalescence).
Without inlet device:
With inlet device:
*Simulations without level control and a simplified eletrocoalescence phenomenon.
Results – without level control
Without inlet device:
With inlet device:
More uniform flow for case with inlet device.
*Simulations without level control and a simplified eletrocoalescence phenomenon.
Results – without level control
Oil volume fraction in outlet plane:
Without gutter Horizontal gutter without vents Horizontal gutter Vertical gutter
• Different configurations of inlet device:
*Simulations without level control and a simplified eletrocoalescence phenomenon.
Definitions: Level controller
• UDF to calculate brine level on measurement
plane (Execute At End);
– Calculate volumetric flow rate at outlet (Define
Profile).
Plane of level measurement
• Brine level obtained from the averaged volume fraction by trigonometric
relations.
LAL
LAH Normal level
Definitions: Eletrocoalescence phenomenon
• Modeling the brine droplets size increase due to the electrocoalescence;
– Phenomenon treated in a simplified way, without solving population balance;
– The droplet size change inside vessel by a step function defined by droplet
position in equipment;
• Two points for droplet size change chosen as they had greater flow uniformity.
Length (m)
Dro
ple
t siz
e (μ
m)
Cases definitions
• 4 cases defined to study droplet size increase and level
control effects;
– Geometry: best result obtained from study 1 Horizontal gutter without vents