International Journal of Scientific & Engineering Research Volume 9, Issue 9, September-2018 380

ISSN 2229-5518
FLOW ASSURANCE IN KUMUJE WET-
`
GAS PIPELINE: ANALYSIS OF PIGGING
SOLUTION TO LIQUID ACCUMULATION
ACCUMULATION

Sarah Akintola, Emmanuel Folorunsho and

Oluwakunle Ogunsakin

Petroleum Engineering, University of Ibadan,

Nigeria
ABSTRACT
Liquid condensation in gas-condensate pipelines in a Conclusively, OLGA simulator proved to be an
pronounced phenomenon in long transporting lines excellent tool in simulating dynamic multiphase flow
because of the composition of the gas which is highly and predicting liquid holdup in wet-gas pipelines in a
sensitive to variations in temperature and pressure hilly terrain.
along the length of the pipeline. Hence, there is a
resultant liquid accumulation in onshore wet-gas KEY WORDS: Wet-gas, slugging, pigging,

IJSER
pipelines because of the pipeline profile. This slugcatcher, pipeline profile, pigs, holdup
accumulation which is a flow assurance problem can
result to pressure loss, slugging and accelerated
pipeline corrosion if not properly handled. INTRODUCTION
Kumuje wet-gas pipeline is an onshore 19" carbon Natural-gas condensate is a low-density mixture of
steel line which is approximately 70 km long in a hydrocarbon liquids that are present as gaseous
hilly terrain with an elevation of 700 m above sea components in the raw natural gas produced from
level. With the pipeline’s maximum design gas many natural gas fields. Some gas species within the
capacity and field operational capacity pegged at 165 raw natural gas will condense to a liquid state if the
and 135 MMSCFD respectively, this study was temperature is reduced to below the hydrocarbon dew
tasked with proposing an efficient pigging scheme for point temperature at a set pressure.
the removal of liquid inventory from the pipeline
using the capacity of the slugcatcher as the basis for Wet Gases
the scheme, also, factors which affected liquid Natural gas that contains significant heavy
accumulation and pigging efficiency was investigated hydrocarbons such as propane, butane and other
using a dynamic multiphase simulator – OLGA. liquid hydrocarbons are known as wet gas or rich gas.
Using OLGA 2016.2, both steady and dynamic runs Wet gas exists solely as a gas in the reservoir
were carried out in other to investigate into some throughout the reduction in reservoir pressure. Unlike
critical factors such as pipe profile and inclination, retrograde condensate, no liquid is formed inside the
pig velocity, gas velocity, bypass pig leakage etc. that reservoir.
influence liquid condensation and holdup in a wet- Liquid drop-out in wet gas pipelines is becoming
gas pipeline. increasingly common because of high changes in the
Of the three (3) pigging schemes considered, scheme composition and low quality of the natural gas
2 proved to be the ideal operational scheme because supply. Predictions of possible locations where liquid
the surge volume (395 m3) generated by the pig is drop-out occurs are, on occasion, very difficult to
within the handling capacity of the slugcatcher (600 obtain. Moreover, estimating the amount of liquid
m3). Also, liquid holdup was seen to be strongly condensation in the gas pipeline is even more
influenced by the pipe profile and a high flow-rate challenging. From an operating gas company
was observed to significantly reduce the volume of prospective, it is fundamental to identify those issues
liquid held-up in the pipeline. and take the appropriate actions to solve them before

IJSER © 2018
http://www.ijser.org
International Journal of Scientific & Engineering Research Volume 9, Issue 9, September-2018 381
ISSN 2229-5518

they significantly affect the operation of the entire cool due to Joule Thompson effect and liquids can
pipeline system (Mark et al., 2016). condense from the gas. Besides pressure and
Flow assurance is a multidisciplinary process temperature, the other factor impacting condensation
designed to prevent pipe blockage and help ensure is the components of the gas composition. The phase
uninterrupted, optimum productivity in oil and gas behavior of wet gas is quite sensitive to pressure,
streams. The fluid journey from reservoir pore to temperature and gas composition. Change in pressure
process facility involves many disciplines using and temperature can condense some of the heavier
advanced technologies. Even long-producing fields molecules in the gas. Thus, the amount of liquid
develop flow assurance problems as time goes by and formed in the pipeline is dependent upon these three
ever-deeper fields bring new challenges that extend (3) parameters and multiphase flow results in higher
the envelope in which the oil and gas industry can friction pressure losses as compared to single phase
safely and economically produce (OTC, 2006). flow (Leksono et. al., 2008).
The main issues considered when designing gas For undulating pipelines with various inclinations
condensate systems are usually pressure drop, liquid and elevations, the gravitational force due to liquids
handling and hydrate prevention. Pipeline pressure must be considered. In wet gas pipelines, liquid
drop is mainly related to selection of correct pipeline holdup is strongly dependent on pipe inclination,
size, while liquid handling relates to slug catcher size especially at low gas velocities. Undulating pipeline
and plant liquid processing capacity. A large profiles reduce the ability of gas to carry or sweep
diameter pipeline will usually give a low pressure liquid in pipeline. It usually occurs in steep segments

IJSER
drop, but a high liquid content, causing liquid of pipe that requires more rates to transport all the
handling problems, while a smaller pipeline diameter liquids up the incline. Due to low velocity, liquid will
will give higher pressure drop, but less liquid content. accumulate at a low spot. Table 1.2 (as presented by
In addition liquid handling and hydrate prevention Rydahi and Shea (2003)) shows the qualitative
are closely tied to the operational procedures of the behavior of undulating pipelines transporting gas and
pipeline, for operations such as rate changes, shut-in liquid at different ranges of gas superficial velocity.
and start-up, blowdown and pigging (OTC, 2006).

Hold Up/Accumulation METHODOLOGY

Liquid holdup HL, is defined as the fraction of an Pigging Operation
element of pipe which is occupied by liquid at same Using OLGA 2016.2 dynamic multiphase simulator,
instant. It is a common phenomenon in two-phase three (3) different pigging scenarios were considered
flow through a vertical pipe; when gas flows at a in order to determine an efficient pigging scheme for
greater linear velocity than the liquid, slippage takes the removal of loaded liquid in the pipeline. Given
place and liquid holdup occurs. Hold up is the cross the complexity of the pipeline profile as indicated in
sectional area occupied by the liquid in the pipe Figure 1, factors which affect pigging efficiency and
carrying the wet gas flow. In multiphase flow, each liquid holdup were investigated using the parametric
fluid moves at a different speed due to different tool available in the software. The pipeline profile
gravitational forces and other factors, with the and properties of the pig used for the simulation are
heavier phase moving slower, or being more held up, highlighted in Figure 1 and Table 1 respectively
than the lighter phase. The holdup of a particular below.
fluid is not the same as the proportion of the total
flow rate due to that fluid, also known as its cut. To
determine in-situ flow rates, it is necessary to
measure the holdup and velocity of each fluid. The
sum of the holdups of the fluids present is unity.
In wet gas transportation, liquid condensation can
occur in systems where the inlet feed conditions to
the pipeline is nominally all in the vapor phase. As Figure 1: Kumuje pipeline profile
the pressure drop occurs in the pipeline, the gas will

IJSER © 2018
http://www.ijser.org
International Journal of Scientific & Engineering Research Volume 9, Issue 9, September-2018 382
ISSN 2229-5518

Table 1: Pig data for simulation model Pigging Scheme 2

PIG DATA The pipeline throughput was kept constant at 120
Type Short Bypass MMSCFD and the outlet pressure decreased from 50
Static Force 19,000 N bara to 45 bara for 3 hours. When total liquid
Wall Friction 9,500 inventory reached stable state, the outlet pressure was
Linear Friction 0 raised back to 50 bara and pigging operation
Quadratic Friction 4,750 conducted.
Mass 600 kg
Pigging Scheme 3
Diameter 19 inches
With the output held at 50 bara, the throughput of the
Leakage Factor 1%
pipeline was raised from 90 MMSCFD to 100
MMSCFD for 16 hours and then the outlet pressure
To determine the efficient pigging scheme for the was decreased to 45 bara before the pig was launched
Kumuje pipeline, the maximum liquid inventory into the pipeline.
method of Yufei et al. (2017) was adopted.
Several factors influence pigging operation, including
RESULTS
topography, diameter, length, OHTC (Overall Heat Pigging Scheme 1
Transfer Coefficient), fluid components, Profile analysis of Kumuje Pipeline prior to pigging
environmental conditions, GOR, gas velocity and so and at a maintained flow rate shows the holdup of

IJSER
forth. It is generally agreed that, surge volume is liquid in relation to the piping angle. The effect of
affected primarily by the velocity of pig and total liquid holdup is more pronounced in the uphill
liquid inventory, which are influenced directly by gas sections of the pipeline where significant hydrostatic
velocity in pipeline (Yufei et. al, 2017). pressure gradients can be induced. Fig. 2 illustrates
It is found that increasing flow rate or decreasing how, for a given low flow rate, liquid holdup varies
operating pressure can raise gas velocity and with gradient In this example, liquid holdup,
diminish total liquid inventory before pigging, while expressed as percentage of cross-sectional area,
boosting the operating pressure or reducing the varies by a factor of 4 for a one percentage point
throughput can lower the gas velocity and slow down change in pipeline gradient.
the pig during pigging operation. Based on the Once holdup has occurred, slugging can be induced
analysis above, considering operability in real field by changes in flow rates or at low points in the
operations and range of operating pressure, three (3) pipeline. When the flow rate in a multiphase pipeline
different pigging schemes were designed by varying is increased, steady-state liquid holdup is reduced.
production rate and operating pressure to determine a Fig. 2 below shows a direct relationship between the
most efficient pigging process which is economically amount of liquid holdup and pipeline profile (angle
feasible and will also produce slugs within the of inclination). Although pipe section 41,175 m to
handling capacity (600 m3) of the slug catcher. 42,973 m has an angle of about 1.61, the liquid
holdup in that section doesn’t show any
Pigging Scheme 1 corresponding increase as compared to other pipe
The throughput of the pipe line was raised from 80 sections, this is because the angle is too steep to hold
MMSCFD to 90 MMSCFD after 5 hours before it a large volume of liquid, hence the relative low
will be increased to 100 MSm3/h. After 15 hours, the volume of liquid holdup.
throughput was increased to 120 MMSCFD, and the
flow rate reduced to 90 MMSCFD before the pig was
launched at the 27th hour. After the pig was received
at the trap, the throughput was increased back to 120
MMSCFD.

IJSER © 2018
http://www.ijser.org
International Journal of Scientific & Engineering Research Volume 9, Issue 9, September-2018 383
ISSN 2229-5518

Figure 3: Liquid Holdup vs. Pipeline angle after 16

hours of simulation (Q16h = 120 MMSCFD)

Prior to commencement of pigging, the flow rate was

decreased to 90 MMSCFD to ensure that the speed of
the bypass pig will be within the allowable limit of 1
– 5 m/s as indicated by Quarini and Shire (2007). At
the 27th hour of the simulation, the pig was launched
and then received at the trap after 11.3 hours. Fig. 4
shows the holdup of liquid in the pipeline after the
pig was removed.
Figure 2: Liquid Holdup vs. Pipeline Angle at a
maintained flow of 80 MMSCFD

In a wet-gas pipeline, the liquid present at a given

location varies with gas flow rate as confirmed in
Fig. 2. At low flow rates, the gas has difficulty
sweeping the entrained liquid along the pipeline and
liquid accumulation (or holdup) occurs. At high flow
rates, however, the liquid is more easily swept along,

IJSER
and liquid holdup is low and of limited impact on the
pressure drop. Figure 4: Liquid Holdup vs. Pipeline Angle after 40
As the gas flow rate is increased, Fig. 3 shows that hours of simulation (Q40h = 120 MMSCFD)
the increased flow resulted in a decrease in the
volume of liquid held-up in the pipeline from a
maximum of 0.37 at 80 MMSCFD to a maximum of Pigging Scheme 1 Efficiency
0.32 at 90 MMSCFD, and 0.3 at 100 MMSCFD. This Vol. of Liquid in Pipe Prior to Pigging (𝐿𝑖𝑞𝑉𝑜𝑙𝑏𝑝 ) =
is because as the flow-rate was increased, some of the 902 m3
liquid became entrapped in the gas and then carried Vol. of Liquid in Pipe after removal of pig at trap
on to the slug catcher. The flow rate was further position (𝐿𝑖𝑞𝑉𝑜𝑙𝑎𝑝 ) = 80 m3
increased to 120 MMSCFD and it translated into the Pigging operation efficiency,
removal of more held-up liquid until steady state was 𝐿𝑖𝑞𝑉𝑜𝑙𝑏𝑝 −𝐿𝑖𝑞𝑉𝑜𝑙𝑎𝑝
𝑃𝑖𝑔𝑒𝑓𝑓 = ∗ 100
reached. Increasing the throughput beyond 120 𝐿𝑖𝑞𝑉𝑜𝑙𝑏𝑝

MMSCFD is not operationally advisable based on the Therefore, 𝑃𝑖𝑔𝑒𝑓𝑓 = 91.1%

design parameters of the pipeline; hence, the line will Although scheme 1 proved to be quite efficient in
have to be pigged to remove the remaining loaded removing loaded liquids in pipe, maintaining an
liquid in the pipeline. operational flow rate of 125 MMSCFD may present
operational challenges since it is close to the
maximum operational rate of 135 MMSCFD.
To determine if the Pigging Scheme 1 is adoptable as
a pigging schedule for the subsequent removal of
loaded liquid in Kumuje pipeline, the amount of
removed liquid was compared with the capacity of
the slugcatcher which is 600 m3. Figure 5 below
indicates the surge volume and the flow rate of liquid
at the exit (Pipe-19 [section 31]) of Kumuje pipeline.

IJSER © 2018
http://www.ijser.org
International Journal of Scientific & Engineering Research Volume 9, Issue 9, September-2018 384
ISSN 2229-5518

From Fig. 6, the liquid hold up in the section is seen

to increase correspondingly with the pipe angle of
inclination as compared with pigging scheme 1 (Fig.
2) which had some exception. This is because the
high gas flow rate and the reduced outlet pressure
allows for liquid to remain in the accumulated pipe
section. Also, the maximum liquid holdup value is
seen to be 0.25 as compared to the 0.37 recorded in
pigging scheme 1.
The estimation of the surge volume for scheme 2
Figure 5: Slug catcher surge volume for pigging shows that maximum surge of 395 m3 is expected at
scheme 1 the slugcatcher after the bypass pig reaches the trap.
From the figure above, it is observed that the This value is within the slug catcher’s design
maximum surge volume is about 654 m3 just as the capacity; hence, this pigging scheme is adoptable for
pig was about to be trapped. This volume exceeds the the removal of loaded liquid from Kumuje gas-
slug catcher capacity of 600 m3 situated at the pipe condensate pipeline.
end; hence, this pigging scheme can’t be adopted
without causing an over-flooding of the slug catcher. Pigging Scheme 2 Efficiency
Vol. of Liquid in Pipe Prior to Pigging (𝐿𝑖𝑞𝑉𝑜𝑙𝑏𝑝 ) =

IJSER
Pigging Scheme 2 534.32 m3
Figure 6 below shows the liquid hold with respect to Vol. of Liquid in Pipe after removal of pig at trap
pipeline profile. The maximum surge volume position (𝐿𝑖𝑞𝑉𝑜𝑙𝑎𝑝 ) = 69 m3
expected at the slug catcher is also indicated in 𝑃𝑖𝑔𝑒𝑓𝑓 = 87.1%
Figure 7.
Pigging Scheme 3
For pigging scheme 3, the maximum liquid holdup in
the pipeline was found to be 0.32. Although this
value is lower than that recorded for scheme 1, the
surge volume is significantly larger than that of
scheme 1. This could be attributed to the higher gas
flow rate in scheme 1 which was sufficient to carry
out a good portion of the liquid before the
commencement of pigging operation.

Figure 6: Variation of liquid hold up with pipeline

profile for pigging scheme 2

Figure 8: Variation of liquid hold up with pipeline

profile for pigging scheme 3
Figure 7: Slug catcher surge volume for pigging
scheme 2

IJSER © 2018
http://www.ijser.org
International Journal of Scientific & Engineering Research Volume 9, Issue 9, September-2018 385
ISSN 2229-5518

0.3
0.25
0.2

Holdp (%)
0.15
0.1
0.05 Liquid
Holdup
0

Gas Velocity (m/s)

Figure 9: Slug catcher surge volume for pigging
scheme 3 Figure 11: Holdup vs. Gas velocity

Pigging Scheme 3 Efficiency Hence, the optimum gas velocity was calculated as
Vol. of Liquid in Pipe Prior to Pigging (𝐿𝑖𝑞𝑉𝑜𝑙𝑏𝑝 ) = 3.46 m/s
1511 m3
Flow Regime
Vol. of Liquid in Pipe after removal of pig at trap
Three (3) different flow regimes were recorded in
position (𝐿𝑖𝑞𝑉𝑜𝑙𝑎𝑝 ) = 75 m3

IJSER
Kumuje pipeline during the dynamic flow simulation
𝑃𝑖𝑔𝑒𝑓𝑓 = 95.0% using pigging scheme 1. The dominant flow in the
pipeline was stratified flow except at pipe bends in
Pig Speed vs. Pig Efficiency
which slug flow is briefly observed. After the pig was
From the simulation result for pigging scheme 1 – 3,
lunched in the 27th hour of simulation, bubble flow
the effect of the velocity on the efficiency of the pig
was briefly observed as indicated in Figure 12.
was determined. Figure 10 below shows that at a
constant leak opening of 1%, the accumulated liquid
removal efficiency of the pig increases with a
decrease in pig speed and this relationship is found to
be valid within a pig velocity of 1 – 5 m/s.

Figure 12: Flow regime in Kumuje pipeline

P&T vs. Liquid Holdup

Pipe section 9 was studied under pigging scheme 2 so
Figure 10: Effect of pig velocity on pig efficiency in as to determine how variations in pressure and
removing loaded liquid temperature affect the condensation of liquid in the
pipeline. Fig. 13 shows the trend profile from OLGA
Gas velocity vs. Holdup dynamic multiphase simulator.

IJSER © 2018
http://www.ijser.org
International Journal of Scientific & Engineering Research Volume 9, Issue 9, September-2018 386
ISSN 2229-5518

temperature over the entire length of the

pipeline.
3. The liquid holdup in the pipeline is strongly
dependent on the pipeline profile and angle
of inclination.
4. There is a direct relationship between pig
velocity and pig efficiency for velocity
values between 1 – 5 m/s.
5. The slugcatcher can be used as a basis for
the design of a pigging scheme
6. Pigging scheme 2 was adopted as the ideal
scheme for the pigging of the pipeline after
it yielded a surge volume less than the slug
catcher’s design capacity.
Figure 13: PT vs. Holdup

Table 2 below presents the summary of the dynamic

simulation of Kumuje gas-condensate pipeline. It can REFERENCES
1. Leksono, M., Evi, W., and Muchammad, I. (2011).
be seen that Pigging Scheme 2 is the most effective
Study to Predict Liquid Holdup in Undulating Pipeline.
way of removing loaded liquid from the pipeline In: IPA, 2011 - 32nd Annual Convention Proceedings,

IJSER
without imposing any additional financial or May 2008.
technical constraint on the operation of the pipeline. 2. Luigi R. (2017). Stratified gas-liquid flow - An analysis
of steady state and dynamic simulation for gas-
condensate systems, Process Simulation Services,
Piave, 17:200, Rescaldina, Italy.
Table 2: Summary of transient simulation of Kumuje 3. Mack, S, Ivor, R., Ellul, H. and William, B. (2016).
Pipeline using OLGA Flow Assurance – At What Cost? Pipeline Simulation
Interest Group (PSIG) 1603.
4. OTC (2006). Flow-Assurance Modeling: Reality Check
and Aspects of Transient Operations of Gas/Condensate
Pipelines
5. Pinto, A., Voss, R. and Ladwa, S. (2017). Operational
Bypass Pigging in Multiphase Lines- A Flow
Assurance Study Comparison to Real Operation and
Model Validation, SPE-188950-MS, pp. 1 – 14.
6. Rydahi, A. and Shea, H. (2003). Liquid Inventory
Management in Wet Gas Pipelines, SPE/
Scanpower Petroleum, 2003.
CONCLUSION 7. Wu, H. and Spronsen, G. (2005). Slug reduction with
Having investigated into flow assurance challenges in high by-pass pigs - a mature technology. In: 12th
International Conference on Multiphase Production
onshore wet-gas pipelines, specifically liquid
Technology, Barcelona (2005).
holdup/accumulation in pipelines; the following 8. Xu, X. and Gong, J. (2005). Pigging Simulation for
conclusions can be drawn after a complete review Horizontal Gas-Condensate Pipelines with Low-Liquid
and analysis of the results obtained from the dynamic Loading. Journal of Petroleum Science and
Engineering, 48(5), pp. 272-280.
simulator:
9. Yufei, W., Renwei, L., Wenguang, W., Xiaying, Du.,
1. Kumuje gas-condensate pipeline model was Zhaoguang, Qu., Chunyu, Liu. and Xin, Q. (2017).
successful built using OLGA dynamic Study on Pigging Solution of Subsea Wet-Gas Pipeline.
multiphase simulator and several simulation In: Proceedings of the Twenty-seventh International
Ocean and Polar Engineering Conference, San
cases were run to study the phenomenon of
Francisco, CA, USA, June 25-30, 2017.
liquid loading in onshore gas pipes.
2. Liquid condensation in wet-gas pipelines is
a function of the sensitivity of the gas
composition to changes in pressure and

IJSER © 2018
http://www.ijser.org