Slug Handling System on NQ Platform A Case Study for CSeries Pipeline Offshore Engineering Services

1
Slug Handling System on NQ Platform - A Case
Study for C-Series Pipeline Offshore
Engineering ServicesONGC Ltd.Mumbai
2
Back Ground

• C - Series fields lie 60 km. West of Daman.
• Comprises of mainly Gas bearing structures
  viz. C-39-8, C-39-6, C-39-1, C-26, C-24, C-23,
  C-22, B-12-7 and B-12-1.
• Water depth variation 19m-30m.
• Each structure has limited production capacity
• Short production life.

3
Facilities to be created

• The Scheme shall be completed in three phases
• Total Unmanned Well Platforms 8 Nos.
• Phase I - 4 ( C-22, C-24, C-39-1 C-39-A)
• Phase I (except pipelines) commissioned in April
  2009
• Phase II (5th Year) - 2 ( C-23 B-12-7 )
• Phase III (11th Year) - 2 ( C-26 B-12-1 )

4
Pipeline network for C-Series
C-39-A
C-39-1
Phase-I
8X 2.5 km
22 X 60 km C-39-A to C-24
Phase-II
C-23
Phase-III
8 X 3 km
C-22
C-24
8 X 2.5 km
8 X 2 km
C-26
B-12-1
C-24 to NQG 28 X 115 km
8 X 3 km
8 X 2 km
B-12-7
Existing
NQG
Proposed
Uran
5
Production Scenario(Peak Rate)

• Gas, MMSCMD Condensate, M3/HR
• Phase I - 3.20
  28.33
• Phase II (5th year) - 3.58
  67.00

6
Brief Description of the Pipeline System

• C-Series pipeline is 175 km. long having diameter
  of 22 28
• The pipeline starts from C-39-A and runs upto
  C-24 with a diameter of 22 and distance of 60
  Km. The line further starts from C-24 and ends at
  NQG with a diameter of 28 and distance of 115
  Km.
• Seabed on the pipeline route slopes down
  gradually from 20 m at C-series end to 60 m at
  NQG end
• The pipeline has been designed to transport 3.58
  MMSCMD gas and 67 m3/hr condensate from the
  marginal fields to NQ process complex.

7
Bathemetry of Pipeline route
8
Slug Flow

• What is slug flow ?
• Slug flow is a two-phase flow pattern which is
  characterized by a sequence of packs of liquid
  separated by long gas bubbles flowing over a
  liquid film inside the pipe and is normally
  associated with high pressure-drops.
• How slug is formed ?
• Slug formation is a phenomenon which is normally
  observed in long distance pipelines having uneven
  terrain (called terrain sluging) and/ or pipeline
  id much higher than that required for normal
  through-put (called hydrodynamic slugging)
• Implications of slug flow on receiving facilities
  
• Unless sized appropriately, the slug flow may
  cause operational imbalance and even cause
  production loss or facilities shut down
  

9
Study Description

• A dynamic simulation of the pipeline was carried
  out through Scand Power Technology, Dubai using
  their Olga Software to ascertain the possibility
  of slugging of the pipeline under various
  operating scenario.
• The objective of the study were to determine the
  slugging of the pipeline under various operating
  scenario like reduced flow, shut down, re-start
  and production ramp-up.

10
Simulation Scenarios

• In order to optimise the number of simulations,
  following typical cases were simulated using Olga
  software.
• Constant production rates Considering 25, 50,
  75 and 100 of maximum production rates for
  Phase-I Phase-II.
• Turn down Simulation Reducing the production
  rates under Phase-II from 100 to 75, 75 to
  50, 50 to 25
• Shut-in Simulation Shutting down the Production
  under Phase-II of each source simultaneously from
  25 to non flowing condition.
• Re-start Simulation Restarting the production
  from non-flowing condition to 25 and non-flowing
  condition to 100.
• Ramp-up Simulation Increasing the production
  from 25 to 50, 50 to 75, 75 to 100 and 25
  to 100.

11
Result of various scenarios simulated

• CONSTANT PRODUCTION RATES
• 1) It is observed that higher the production
  rate, lower is the liquid content in the
  pipeline.
• Mild terrain slugging was observed at upstream
  ends i.e. well heads, for 25 production rate
  for both, Phase-I and Phase-II. Maximum liquid
  content observed in the pipeline ( 2764 m3 in
  Phase-I and 3295 m3 in Ph-II).
• No slugging was observed at the downstream end at
  NQG and fluid arrival rate at NQG was found to be
  well within the handling capacity of NQG topside
  facilities.
• 4) Slug flow resulted in well head pressure
  fluctuations of approx. 1 bar. The slug frequency
  was approx. 1 slug per day for Phase-I and 2.8
  slugs per day for Phase-II.
• Contd.

12
Result of various scenarios simulated (contd.)

• TURN DOWN SIMULATION
• It is observed that higher the turndown, higher
  is the time for liquid content to get stabilized.
• 265 hrs. (11 days) are required for liquid
  content to get stabilized in 28 C-24-NQG
  pipeline when production rate is reduced from
  100 to 25. Whereas stabilization takes around
  70 hrs. (3 days) when production rate is reduced
  from 100 to 75.
• No slugging was observed during the stabilization
  period for any turndown case.

13
Result of various scenarios simulated (contd.)

• SHUT IN SIMULATION
• It is observed that after the wellheads were
  shut-in for 11.5 hrs, the riser base towards NQG
  was completely filled up with liquid due to down
  slope geometry of the pipeline towards NQG.
• After 24 hours shut-in, the liquids had
  completely filled section from KP 110.63 to riser
  base having pipe volume 175 m3. Moreover, the
  pipe from KP 108.85 to KP 110.63 was partially
  filled accounted for 273 m3 liquid from KP 108.85
  to the riser base.
• The liquid accumulation towards riser base at NQG
  gives rise to slug upon restart. It was also
  observed that higher the period of shutdown,
  higher is the volume of slug upon restart.

14
Result of various scenarios simulated (contd.)

• RESTART SIMULATION
• 1) Both 100 and 25 restart resulted in slug
  flow and were beyond the NQG handling capacity.
  Reduced restart flow rate does not have much
  impact except prolonging the arrival rate of
  slug.
• 2) Liquid slug was also followed by gas surge of
  magnitude much higher than the design capacity of
  the topside facilities.
• 3) For restart to 100 production case, the
  liquid rate surged to a peak value after 3 hours
  had passed since flow was resumed. The total
  liquid surge volume in excess of handling
  capacity (100 m3/hr) was 269 m3 and the
  processing time required was 3 hrs. A gas surge
  of magnitude 16 MMSCMD (for a short duration)
  followed by liquid slug of 12758 m3/hr (for a
  short duration) was also observed. Subsequent
  surges were found to be within the handling
  capacity.
• Contd.

15
Result of various scenarios simulated (contd.)

• RESTART SIMULATION
• For restart to 25 production case, the liquid
  rate surged to a peak value after 9.3 hours had
  passed since flow was resumed. The total liquid
  surge volume in excess of handling capacity (100
  m3/hr) was 286 m3 and the processing time
  required was 3.5 hrs. A gas surge of magnitude 11
  MMSCMD (for a short duration). followed by liquid
  slug of 16688 m3/hr (for a short duration) was
  also observed. A subsequent gas surge of equal
  magnitude was observed at10th hour.
• It is also observed that any shutdown greater
  than 1 hour will produce slug on restart.
  However, additional restart simulation is
  required to quantify the same.
• Contd

16
Result of various scenarios simulated (contd.)

• RESTART SIMULATION
• 6) Experienced velocities exceeding erosional
  velocity ratio by a factor of 1.1 in C-24 22
  riser during 100 restart operation. However,
  erosional velocity limit was not exceeded in the
  pipeline network.
• 7) Experienced velocities exceeding erosional
  velocity ratio by a factor of 1.4 in C-24 22
  riser during 25 restart operation. The average
  velocity did not exceed the erosional velocity
  limit but due to slug flow, the gas velocity
  instantaneously surged beyond the erosional
  velocity.

17
Result of various scenarios simulated (contd.)

• RAMP-UP SIMULATION
• Ramp-up in steps i.e. increasing the production
  from 25 to 50, 50 to 75 and 75 to 100
  allowing the flow to stabilize after each
  increment do not cause any slug flow at NQG. Such
  ramp-up from 25 flow rate to maximum production
  rate could be achieved in 3 days.
• The liquid surge for 25 to 100 ramp-up was
  found to be greatest volume of ramp-up surges
  analyzed. The liquid surge volume produced was in
  excess of NQG handling capacity.
• During 25 to 100 ramp-up, erosional velocity
  limit was exceeded for few segments of pipeline
  network.
• During sequential ramp-up, erosional velocity
  limit was only exceeded in C-24 22 riser.

18
Recommendations for possible remedial measures

• Increasing handling capacity leading to reduction
  in time required to process liquid surges.
• Implementation of topside choke leading to
  reduction in surge volumes within the current
  vessel volumes and handling capacity rates.

19
Implementation of the Scheme

• Since the platform facilities have already been
  put in place, any addition of facilities for
  increasing capacity would involve time and cost.
  This would result in delay in commissioning the
  plant or shutdown for long duration.
• Containing the surge flow at the inlet with choke
  / control valve for the duration that the surge
  is anticipated is most reasonable solution to the
  problem.
• The PID (Piping and Instrumentation Diagram) of
  the process before the slug control scheme is
  implemented is shown in the next slide.

20
Implementation of the Scheme (Before Slug
Control)

21
Implementation of the Scheme

22
Implementation of the Scheme (contd.)

• A combination of the shutdown valve and control
  valve on the bypass line would ensure safe
  operation during the start-up of the line after a
  shutdown.
• 6 control valve has been found suitable for the
  control of the slug flow and intermittent gas
  flow. Any lower size would become a constraint
  for the quantity of the gas flow envisaged
  intermittently during the slug flow.
• In addition, inlet shutdown valve is provided
  upstream of control valve to shutdown facilities
  during for upset condition during slug flow.

23
Implementation of the Scheme

• The 6 SDV shall be opened manually for START-UP
  at site. (The control valve is kept crack open
  during this activity).
• The 24 SDV will remain in closed condition
  during slug flow until it is safe to open for
  normal operation once the slug is cleared.
• To ensure above, independent local pneumatic
  selector pull-in / reset type 3-way selector
  valve is provided for both SDVs.
• Once the slug flows ceases, 24 SDV shall be
  opened and then 6 SDV shall be closed.
• Schematic for manual selection of the SDV is
  shown in next slide.

24
Implementation of the Scheme Control Schematic

25
Implementation of the Scheme Control Valve

• The 6 Control Valve has a Hand wheel for on
  site manual operations of the valve if necessary.
  
• Operation of the control valve can be done from
  control room, using the Controller being provided
  for this valve. Manual or Auto operation mode is
  operator selectable.
• To begin with, the operation of start up is
  envisaged to be manual with a pre determined
  Crack opening (up to 20 opening). Once the line
  pressure starts dropping from 26 kg/cm2g to 11
  kg/cm2g, the valve may be opened further
  gradually over a period of time and level of KOD
  downstream is continuously monitored alongwith
  inlet pressure.
• The KOD has its own level controller which shall
  maintain the KOD level. When there is liquid
  inflow (Slug) which is more than the valve can
  handle, (i.e. LCV full open), the operator shall
  start to close the SCV gradually to reduce inflow
  and maintain the KOD level.

26
Implementation of the Scheme Control Valve
(contd.)

• FEEDBACK CONTROL METHODS
• SSIC shall keep the SCV sufficiently open to
  allow 80 M3 /Hr flow to maintain KOD level
  stable. This opening () will be known better
  with experience over a period of time.
• The level controller (LIC) for KOD level control
  (Condensate level) is utilized for determining
  the operation of the SCV in feed back mode. LIC
  output shall become PV (Process Variable) for
  the SSIC. Set point of SSIC shall be set at a
  value in the range 75 to 80 (i.e. 16 to 17mA in
  4-20mA). Any increase in the PV value of SSIC
  beyond this shall start to close SCV.
• SSIC shall keep SCV open for all PV values up-to
  the SET-POINT value.

27
Implementation of the Scheme Control Scheme

28
Implementation of the Scheme Control Valve
(contd.)

• Alternately, LT output (KOD level) can be used
  as PV (Process Variable) for SSIC. Set point is
  set at a value in the range 75 to 80 (i.e. 16mA
  to 17mA in 4-20mA). Any increase in the PV value
  of SSIC beyond this shall start to close SCV.

29
Conclusion

• It is observed that the adopted method of slug
  control for maintaining KOD level is most
  appropriate as it is not dependent on complicated
  software which use assumed conditions and
  predictive methods for slug control.

30
Thanks