SPE 122829 Slug Catcher Conceptual Design as Separator for Heavy Oil J. Marquez, C. Manzanilla, and J. Trujillo, PDVSA Intevep
Copyright 2009, Society of Petroleum Engineers This paper was prepared for presentation at the 2009 SPE Latin American and Caribbean Petroleum Engineering Conference held in Cartagena, Colombia, 31 May–3 June 2009. This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.
Abstract The gas-liquid separation processes in the heavy and extra heavy oil fields are performed mainly with gravity conventional separators. However, the separation efficiency of these equipments depends on the operating conditions, the appropriate design and the properties of the fluids. Thus, in heavy and extra heavy oil fields, the separation efficiency is affected due to of the handle liquids of viscosity high, low pressures and low gas flow rate. Additionally, these conditions increase the probability of slug flow formation through pipelines that causes operational problems, mainly in the separation process. In response to this situation, arises the need to design a gas-liquid separator able to handle viscous liquid, reduce the effects of slug flow and perform an efficient separation. In this sense, there are different technologies that can help to improve the gas-liquid separation, among them is found the finger type slug catcher. This technology is usually used in gas fields or light and medium oil fields as conditioner flow and slug flow mitigator. However, this paper has changed this focus for considering a design for heavy oil. This paper presents an improvement in the methodology of Sarica et al. (1990) to predict the dimensions of finger type slug catcher to use in heavy oil fields. It is based on the effect from the transition of stratified flow to non-stratified flow when the liquid phase is viscous and only considers slug flow characteristics under normal flow. Based on the improvement the required diameter and length of the finger are determined. The improvement is used to design a finger type slug catcher for heavy oil field conditions in Orinoco Belt and an economic comparison against conventional separators is presented. The comparison demonstrated that the finger type slug catcher designed with the proposed method is less expensive than the conventional separator. Introduction The multiphase flow through pipelines in heavy oil fields is common everyday. In this scenario the slug flow pattern is promoted due to the high oil viscosity, the low pressure and the low gas flow. In fact, the envelope of slug flow in gas-liquid flow pattern transition maps is increased with the increase of the liquid viscosity or the liquid flow rate. Pipelines transport the production streams to the processing facilities where the separator is the first device to receive them. Separators could be affected by severe operational conditions caused by slug flow. Therefore, it is necessary to use a gas-liquid separator capable to mitigate the slug flow effects and perform an efficient separation. Although several gas-liquid separation technologies have been available for a long time, a study was led to identify the more suitable technologies for heavy oil applications. The study found that the conventional separator is widely used in heavy oil field developments. Also, there are pre-separators which can be used as flow conditioners to improve the separation. However, a pre-separator could be also used as a primary separator. Among the found technologies there are conventional separators, slug catchers, “T” junctions, ultrasonic equipments, etc. Among these technologies is selected the finger type slug catcher to make a conceptual design that considers the viscous liquid effects. The traditional design of conventional separators is based on a typical residence time depending on the oil density and not considering the viscosity liquid effects on the gas bubble rise velocity entrained in the liquid. Even considering the theory of
2
SPE 122829
the bubble velocity there are simplifications that do not reproduce accurately the hydrodynamic behavior of the gas-liquid separation into the equipment when the liquid is viscous. The desired hydrodynamic behavior into a gas-liquid separator is achieved when the gas and liquid phases are stratified. But if slug flow arrives to the separator, the ideal situation is achieved when the effective area of flow is increased enough to reduce the velocity and mitigate the slug. Thus, the inviscid Kelvin-Helmholtz instability criterion (IKH) allows identifying if stratified flow can exist under the given conditions, including the pipeline geometry. Based on the IKH criterion that was used by Taitel and Dukler, (1976) which is widely used in the oil industry to predict the stratified non-stratified transition, Sarica et al. (1990) proposed a design methodology for sizing a slug catcher. However, according to Lin and Hanraty (1986) and Barnea (1990) the IKH criterion does not properly predict the transition when the liquid phase is viscous. However, Barnea and Taitel (1993) proposed a criterion called viscous Kelvin-Helmholtz instability (VKH) that considers the effects of the shear stress but not the interfacial tension effects, i.e. the viscous effect is considered. This paper proposed an improvement in the original methodology of Sarica et al. (1990) to design the finger type slug catcher. The improvement considers the use of the viscous Kelvin-Helmholtz instability (VKH) criterion proposed by Barnea and Taitel (1993). Thus, the diameter of the finger type slug catcher is determined to guarantee stratified flow through the device. This methodology was applied to design a slug catcher for heavy oils and an economic comparison against conventional separators is presented. METHODOLOGY OF DESIGN The proposed conceptual design of finger type slug catcher for heavy oil is based on the methodology proposed by Sarica et al, (1990). Then, the design required information related to the characteristics of slug flow at the pipelines conditions of inlet to the slug catcher. Also, it must be calculated the liquid accumulation and finally obtain the main dimensions of the equipment, such as diameter and finger length. The methodology assumes in the same way that Sarica et al. (1990) the next: • During the production of the liquid slug, some liquid is shed to the liquid film of the Taylor bubble due to the difference of velocities between the slug and Taylor bubble. For this reason, the volume of liquid accumulation is less than the calculated. • It is supposed that prior to the liquid slug production, the operational holdup is at the minimum, i.e. that the liquid level into the catcher will fall during Tailor bubble production. These two assumptions give a safety factor to the design of the equipment that slightly increases the slug catcher dimensions. Other consideration is related to the fact that the fingers must be in horizontal position because negative or positive inclinations generate waves of high amplitude that produce early liquid carryover. Additionally, under this condition the criterions to predict the stratified-no stratified transition do not work properly (Barnea. D., 1990; Barnea, D. and Taitel, Y., 1994). The methodology is given for one finger, but can be adapted to more than one finger if the liquid distribution among the fingers is considered. Slug flow characterization In this section it is presented how is predicted every characteristic of slug flow in order to obtain the proper design of the slug catcher. These characteristics include slug holdup, slug length, slug velocity, translational velocity, gas and liquid velocity at Taylor bubble, gas and liquid velocity for the liquid slug and slug frequency. The slug holdup is determined according to the liquid viscosity. If the viscosity is less than 500 cP the holdup is calculated using the Gregory et al. (1978) correlation as,
H LLS =
1 1.39
⎛ v ⎞ 1+ ⎜ M ⎟ ⎝ 8.66 ⎠
(1)
For viscosity greater than 500 cP the holdup is determined using a correlation obtained at PDVSA Intevep as,
H LLS = 1.0046 ⋅ e −(0.0022⋅ReL )
(2)
SPE 122829
3
The holdup in Taylor bubble is determined using a correlation obtained at PDVSA Intevep as,
H LTB =
(vTB − v L ) ⋅ H LLS
(3)
vTB
The gas void fraction is expressed using the Beggs (1991) correlation
α S = 1 − H LLS
(4)
The film length of the slug unit is predicted using a correlation developed with high viscosity liquid experimental data at PDVSA Intevep.
⎛ Re SL L F = 0.0365 ⋅ ⎜⎜ Re ⎝ SL + Re SG
⎞ ⎟⎟ ⎠
−0.8606
(5)
The slug length correlation is obtained by inserting LF into the correlation developed by Shoham (2000) to predict the film length. This correlation only considers hydrodynamic slug flow. ⎛ Re SL 0.0365 ⋅ ⎜⎜ Re ⎝ SL + Re SG LS = vL ⋅ H LLS −1 vSL
⎞ ⎟ ⎟ ⎠
−0.8606
(6)
The correlations already presented require the estimation of velocities, such as: mixture velocity, liquid superficial velocity, and gas superficial velocity that are easy of obtaining. Also, the prediction of the translational velocity and the drift velocity is necessary. The translational velocity is given by the Nicklin (1962) correlation and depends on the velocity profile, mixture velocity and drift velocity:
v TB = Cv M + v D
(7)
Where C is given according to the flow type: If the flow is laminar: C = 2 , while if it is turbulent C = 1.2 and when it is between laminar and turbulent the Taitel (2000) correlation is used. C=
2.0 ⎛ Re L 1 + ⎜⎜ ⎝ Re CL
⎞ ⎟⎟ ⎠
2
+
1.20 ⎛ Re 1 + ⎜⎜ CL ⎝ Re L
⎞ ⎟⎟ ⎠
2
(8)
The drift velocity is calculated depending on the pipe inclination, such as: For pipes slightly inclined the drift velocity is estimated by Bendiksen (1984) correlation as, v D = (v D )horizontal ⋅ cos (θ ) + (v D )vertical ⋅ sin (θ )
(9)
The following correlation is used for horizontal pipes
(vD )horizontal
= 0.54 gD
(10)
And for vertical pipes the following correlation is used.
(vD )vertical = 0.35 The slug frequency is estimate as,
gD
(11)
4
SPE 122829
fS =
vTB LU
(12)
Other parameters of interest for slug flow characterization are the liquid and gas instantaneous flow at the inlet of the catcher. They are calculated using the Miyoshi et al. (1988) model. For the liquid,
QinsL = v Mins A p H LLS
(13)
QinsG = v Mins A p (1 − H LLS )
(14)
And for the gas,
Prediction of Liquid accumulation
The liquid accumulation into the slug catcher can be estimated applying a liquid mass balance between the inlet and outlet of the equipment (Sarica et al. 1990) as, ⎡Liquid input ⎤ ⎡Liquid discharger ⎤ ⎡Liquid accumulation ⎤ ⎢mass rate ⎥ − ⎢mass rate ⎥ = ⎢mass rate ⎥ ⎣ ⎦ ⎣ ⎦ ⎣ ⎦
(15)
Where the liquid input mass rate is determined through the Miyoshi et al. (1988) model to calculate the liquid instant flow. While the liquid discharge mass rate is related at the liquid flow to the outlet of the catcher. And this liquid flow depends on the flow control valve size. Based on the liquid mass balance presented by Sarica et al. (1990), the accumulated liquid volume is given as,
Vaccum = t sp * Qacum =
[
LS max v M H LLS A p − Qdis vTB
]
(16)
Finger type slug catcher dimensioning
The most important parameter in the slug catcher design is the diameter of the slug catcher fingers, which is calculated to obtain stratified flow into the finger. In this sense, models to predict the transition from slug flow to stratified flow are necessary, such as: the inviscid Kelvin-Helmholtz instability criterion (IKH), the viscous Kelvin-Helmholtz instability criterion (VKH), Taitel and Dukler (1976) model, etc. To predict the transition Sarica et al. (1990) used the inviscid Kelvin-Helmholtz instability criterion (IKH) presented by Taitel and Dukler (1976). However, in this work is proposed to use the viscous Kelvin-Helmholtz instability (VKH) criterion presented by Barnea and Taitel (1993) to determine the transition from slug flow to stratified flow because it predicts better the transition for a wider range of viscosities (100-5000 cP): The criterion is expressed as,
v Gtran
⎡ ⎢ ⎛ ρ − ρG ≥ K V ⎢(ρ L RG + ρ G R L )⎜⎜ L ⎢ ⎝ ρ L ρG ⎢ ⎣
⎤ ⎞ AP ⎥ ⎥ ⎟g ⎟ ∂A ⎥ L ⎠ ∂h L ⎥⎦
In this expression, KV is a correction factor given as,
1/ 2
(17)
SPE 122829
5
KV = 1 −
(CV
− C IV )
2
ρ L − ρG A g ⋅ cosθ ⋅ P dAL ρ
(18)
dhL The VKH criterion provides the minimum diameter from which can be obtained stratification, i.e. it works on the transition curve, such as is shown the green point at the Fig. 1.When the actual gas velocity is less than the transition gas velocity is expected the stratified flow. Thus, the catcher diameter should be bigger than the minimum diameter to receive the incoming liquid. The catcher diameter is determined increasing the minimum diameter to insure stratified flow into the equipment, such as is shown in the Fig 1 as operation point. Also, it must consider the incoming liquid flow, available space for installation and the costs.
100
10
VsL (m/s)
Intermittent 1
Annular 0.1
Stratified
0.01 0.1
1
10
100
VsG (m/s)
VKH
HL/D = 0.5
Operation point
Transition point
Fig. 1. Flow partner map for the designed slug catcher. For a given gas superficial velocity there is a transition liquid holdup and an operation liquid holdup. The first is given by the maximum liquid superficial velocity for stratified flow and is calculate using the VKH criterion. The second is given by the average operation flow rates of liquid and gas at the slug catcher. The difference between these two holdups will provide the available volume to handle the accumulation of liquid in the slug catcher. Thus, the catcher length for the designed diameter is given as,
L finger =
A finger
Vaccum H L trans − H L oper
[
]
(19)
6
SPE 122829
RESULTS AND DISCUSSION
The proposed methodology is used to design a finger type slug catcher for a heavy oil field in Orinoco Belt. The designed equipment was economically compared with a design of a conventional horizontal separator that is commonly used in the field. The data for the design are given in Table 1 and the location of the operation point in the flow pattern map for this information is shown in the Fig 2. Table 1. Field data º API 16 Temperature (º F) 72 – 95 Pressure (psig) 105.80 QL (BPD) 14.343,55 QG (MMSFD) 8,556 BS&W (%) 42,65 γG 0.55
100
VsL , (m/s)
10
Intermittent
1
Annular 0.1
Stratified 0.01 0.1
1
10
100
VsG (m/s) VKH
HL/D = 0.5
Operation point
Fig. 2. Flow pattern map for the inlet conditions of the catcher. In the catcher design four fingers were considered and the distribution of flow rate through of finger is uniform. Under these considerations the catcher diameter and length are determined. Thus, the diameter is 0.508 m and the length of every finger is 8 m. The weight of the catcher considering every pipe section is 3573 kg. The operation point of the finger is showed in the Fig. 1 as the blue point. On the other hand, a horizontal conventional separator for the same conditions will have approximately a diameter of 1.83 m (6 ft) and a length of 6.10 m. (20ft).The separator weight is 4627 kg. Based on to the weights of the equipments, the type finger slug catcher is 23% lighter than the horizontal conventional separator. For this reason, it is expected a similar reduction in the fabrication cost. Whenever is considered that the other cost relating to the fabrication are almost the same. CONCLUSIONS.
The slug flow characteristics must be known to develop a proper design of finger type slug catcher. Thus, various correlation and models are selected to predict the slug flow characteristics for heavy oil in a rigorous way.
SPE 122829
7
It is proposed a improvement of the Sarica et al. (1990) methodology based on the use of the viscous Kelvin-Helmholtz instability (VKH) criterion to predict the stratified-no stratified transition in a more rigorous way to determine the dimensions of a finger type slug catcher to handle viscous liquids. The improvement allows performing a better design of the slug catcher to guarantee the segregation and separation of the phases while the slug mitigation is achieved. The economical comparison based on the weights demonstrated that the fabrication cost of the slug catcher can be 23% less expensive than the conventional separator. REFERENCES
1. 2.
Barnea, D., Taitel, Y., 1994. Interfacial and structural stability of separated flow. Int. J. Multiphase Flow 20: 387-414. Barnea, D., y Taitel, Y., 1993. Kelvin – Helmholtz Stability Criteria for Stratified Flow: Viscous Versus Non-Viscous (Inviscid) Approaches. Int. J. Multiphase Flow. 19(4): 639-649. 3. Barnea. Dvora., 1990.On the effect of viscosity on stability of stratified gas – liquid Flow – application to flow pattern transition at various pipe inclinations. Chemical Engineering Science, 46(8): 2123-2131. 4. Beggs, H.D., 1991. Production Optimization Using NODALTM Analysis. OGCI Publications Oil & Gas Consultants International Inc. Tulsa. 5. Bendiksen, K. 1984. An Experimental investigation of the motion of long Bubbles in Inclined Tubes, Int. J. Multiphase Flow 13(1): 1-12. 6. Lin P. and Hanratty, T. 1986. Prediction of the initiation of slugs with linear stability theory. Int. J. Multiphase Flow 12: 79-98. 7. Miyoshi, M., Doty, D. and Schmidt., 1988. Slug – Catcher Design for Dynamic Slugging in Offshore Production Facility. JGC Corp. SPE 14124. 8. Nicklin, D. 1962. Two phase bubble flow, Chem. Eng. Sci, 17: 693-702. 9. Sarica, C., Shoham, Ovadia., and Brill, J.P., 1990. A New Approach for Finger Storage Slug Catcher Design. OTC 6414. 10. Shoham, O., 2000. Two-Phase Flow Modeling. Department of Petroleum Engineering. University of Tulsa. TOMO 1. 11. Taitel, Y. and Dukler, A., 1976. A Model for Predicting Flow Regime Transitions in Horizontal and Near Horizontal Gas – Liquid Flow. AICHE J.,22,:47-55 NOMENCLATURE °API API gravity A Cross-sectional area [m2] C Wave velocity fs Slug frequency [slugs/s] g Gravity acceleration [m/s2] H Holdup Kv Coefficient of stability L Length [m] Q Flow rate [m3/s] Re Reynolds number t Time [s] Velocity [m/s] ν
V BS&W
Volume [m3] Bottom sediment and water
Subscripts
accum D dis. F G ins IV
Accumulation Drift discharge Liquid film (Taylor bubble) gas instantaneous inviscid
8
L LS M max oper p S sG sL sp TB trans U V
SPE 122829
liquid Liquid slug Mixture gas-liquid Maximum Operational Pipe Slug Superficial gas Superficial liquid Slug passage Translational Transition Slug unit Viscous
Greek Letter
γ
specific gravity
θ
Inclination angle
ρ
Density [kg/m3]
