Operational Procedures for Managing High-Pressure Flow Assurance in Complex Subsea Tie-Backs

Why high-pressure tie-backs punish small operating errors

A long subsea tie-back with high shut-in pressure leaves very little room for improvisation. The flowline runs through cold seawater for its entire length. Any unplanned shutdown, slow restart or inhibitor shortfall can push the fluid into the hydrate-forming envelope, and high pressure widens that envelope. Once a hydrate plug forms in a subsea flowline, conventional intervention can't reach it. Remediation means days to weeks of lost production, and the depressurisation it usually calls for carries real safety risk. The Rodanco overview also flags wax and scale as threats that sit alongside hydrates in long tie-backs, and each one opens another path to restricted flow.

This article is about operating practice: what production engineers should have in place before anything goes wrong, how to run transients, and how to keep chemical delivery and flow efficiency healthy for the life of the field.

Standards and requirements context

Few standards spell out flow assurance operating practice directly. Most of it lives in project-specific operating philosophies, which matters because a hydrate strategy that relies on inhibitor injection is only as good as the injection path behind it.

• ISO 13628 (Petroleum and natural gas industries — Design and operation of subsea production systems) is the series covering subsea production system design and operation. Use the relevant part for the control system and equipment you operate.
• IEC 61511 (functional safety for the process industry sector) governs safety instrumented functions. It applies if the tie-back depends on high-integrity pressure protection or topside shutdown functions to protect flowlines and risers.

The governing document in practice is your field's flow assurance operating strategy, which comes out of the design-phase analysis. The Bijupira and Salema paper (OTC, 2004) lays out this route: flow assurance analysis done specifically to develop operational strategies for deepwater developments. If that strategy never gets turned into procedures operators can actually carry out, it protects nothing.

Threats and where they sit in the geometry

Map the hazards onto the layout

The Rodanco overview describes the pattern. The same line can face hydrates when it cools, wax when temperature drops below the wax appearance point, and scale where waters mix. Each hazard concentrates in different places:

• Long cold flowlines: cooling during shutdown and low-rate operation.
• Low points and sags: liquid and water accumulation, which dilutes inhibitor and gives plugs a preferred place to form.
• Risers: severe slugging, elevated liquid holdup and large pressure swings.
• Jumpers, manifolds and dead legs: stagnant fluid that cools first and is least likely to see inhibitor.
• Commingling points: scale and incompatibility risk where different produced waters meet.

Keep an annotated profile of the system that marks these locations and ties each one to the procedure that protects it. Operators who know where the cold, stagnant and water-prone sections are make better calls during an upset.

Steady state is not the problem; transients are

A tie-back sitting comfortably outside the hydrate region at normal rate can slip into it within a single shutdown cycle. The operating risk clusters around four situations:

1. Planned and unplanned shutdowns, when fluid cools toward seawater temperature.
2. Restarts, when cold fluid and liquid inventory get displaced.
3. Turndown, when lower velocity increases cooling and liquid accumulation.
4. Rising water cut, which changes inhibitor demand and hydrate risk.

Shell's work used dynamic simulation to support commissioning and operation of a long gas-lifted tie-back. It's a good model for any complex tie-back: simulate the transient before you perform it.

Hydrate prevention strategies compared

The right mix depends on geometry, fluid, and the topsides and umbilical capacity you have. The qualitative comparison below is useful when you're reviewing or revising a strategy.

In practice these strategies get layered. Insulation stretches the cooldown time, and inhibitor or depressurisation then covers the longer shutdowns. The operating procedure has to say which layer applies in each situation, and who authorises moving between them.

Treat the inhibitor system as a critical barrier

Once inhibitor injection becomes a primary barrier, API 17TR5 stops being just a design reference and becomes operations guidance too. Blockages in chemical injection lines and control systems can come from the chemicals themselves, from contamination, or from incompatibility between fluids. For operators, that means:

• Keep chemicals clean and filtration under control at the topsides injection skid.
• Qualify any new chemical, or any change of supplier or formulation, for compatibility with the other fluids and materials it will meet, before it goes into the line.
• Trend injection pressure and flow for each line, so partial restriction shows up early rather than at the next shutdown.
• Know which injection points are redundant and which are single-path. Plan shutdown and restart around the single-path ones.
• Don't leave chemicals sitting stagnant in lines at conditions where they might degrade or deposit.

Inhibitor demand follows the water phase, so tie dosing to measured or best-estimate water production. Dosing against a stale water-cut figure is a common way under-inhibition goes undetected as a field matures.

Running transients: shutdown, restart and gas lift

Standardise through integrated procedures

Offshore Magazine reports on a project that relied on a multi-disciplinary team. The lesson for any complex tie-back is that flow assurance can't live in a document held by one discipline. Procedures such as shutdown preservation, restart, inhibitor changeover and pigging cut across subsea, topsides, process, control room and chemical supply responsibilities. Write them as single integrated procedures, with a clear owner for each step.

Shutdown preservation

A good shutdown procedure answers these questions before the event, not during it:

1. What is the expected time before the system enters the hydrate region, based on the latest simulation or cooldown data?
2. Is inhibitor injected into the flowline, trees and jumpers, and which valve sequence displaces the fluid?
3. When does the decision to depressurise or displace become mandatory?
4. Who can authorise restart, and what conditions must be confirmed first?

Cooldown time moves with ambient conditions, flowline condition and fluid state, so derive these thresholds from your own field model and record them in the approved procedure. Don't copy them from another field.

Restart

Restart is where liquid inventory, cold fluid and pressure surges all arrive together. Before introducing flow:

• Confirm inhibitor is where it should be and that injection is actually working.
• Define the ramp-up profile to manage slug arrival at the riser and separator.
• Brief the control room on the expected pressure and temperature responses, and on what to do if the response deviates.

For gas-lifted tie-backs, the 2009 Shell paper's emphasis on dynamic simulation is instructive. Gas lift interacts with flowline hydraulics, and unsteady behaviour during startup can look like a fault when it's really a predictable transient. Simulating the sequence beforehand lets operators tell normal from abnormal.

Maintaining flow efficiency over field life

Flow efficiency drops gradually, through deposition, rising liquid holdup, partial blockage and shifting fluid composition. Keep these routines going:

• Trend flowline differential pressure against rate and fluid conditions, and investigate any sustained drift from the commissioning baseline.
• Reconcile production data against model predictions on a regular cycle, and update the model when fluid properties change.
• Run pigging at the planned frequency in the procedure, and review returns for wax, scale or debris.
• Review slugging behaviour at the riser whenever rates, gas lift or choke settings change.
• Reassess the hydrate and wax envelopes whenever new fluid samples come in.

Hydrate plug response: safety first

A suspected hydrate formation is a safety event first and a production event second. Any remediation that depressurises, opens or works on hydrocarbon-containing equipment has to follow your approved well-control and process safety procedures. As a minimum, cover the following:

1. Isolation: positively isolate the affected section and identify all trapped-pressure volumes.
2. Depressurisation: plan controlled depressurisation, from both sides of a suspected plug where the strategy allows it. A plug under differential pressure can move violently.
3. Verification of zero energy: confirm zero pressure by independent means before breaking containment. Don't trust a gauge reading in a blocked line.
4. Lock-out/tag-out: apply LOTO to all isolation points and energy sources.
5. Hazardous-area precautions: control ignition sources and use certified equipment in classified areas.
6. Gas detection: use fixed and portable detection in the work area, and confirm it's working before venting.
7. Safe venting: route released gas and liquids to the flare or vent system designed for them, never to atmosphere in an uncontrolled way.

Hydrate dissociation also releases gas, so the topsides system has to be able to handle the volume released. Bring in the flow assurance specialist and the facility's process safety lead before you start.

Illustrative scenario: unplanned shutdown with rising water cut

This is a constructed scenario for guidance and does not describe a real event.

A tie-back produces through a long cold flowline and a riser. Water production has climbed over several months, but inhibitor dosing hasn't been adjusted since the last review. Then a topsides trip stops flow.

The shift team opens the shutdown procedure. The first question is whether the preservation plan still matches the current water cut. It doesn't. The plan was written for an earlier, drier fluid. The team runs the field cooldown model with the updated water cut, and it shows less time before the system enters the hydrate region than the procedure assumed. Operations leadership decides to inject inhibitor into the vulnerable sections straight away, rather than waiting for the cause of the trip to be fixed.

On restart, trend data from the injection lines show one path behaving abnormally. Following API 17TR5 principles, the team treats this as a possible partial blockage, switches to the redundant path, and plans an investigation. Restart is staged, and the control room is briefed on the expected slug arrival at the riser.

Two lessons come out of this. The procedure's assumptions have to be tied to current fluid data, and injection-line trends are a leading indicator that needs watching all the time, not just at shutdown.

Practical checklist

• Is there an annotated flowline profile showing low points, dead legs, risers and commingling points?
• Does each shutdown, restart and turndown procedure state which hydrate strategy layer it relies on?
• Are cooldown and inhibitor-demand assumptions updated for current water cut and fluid properties?
• Is every injection line trended, with single-path lines identified?
• Have new or changed chemicals been checked for compatibility before injection?
• Has restart been simulated, including slug and gas lift behaviour where relevant?
• Are procedures integrated across subsea, topsides, control room and chemical supply, with named owners?
• Is a plug-response procedure approved, covering isolation, depressurisation, zero-energy verification, LOTO, gas detection and safe venting?
• Is a pigging plan in place, with a stuck-pig contingency?
• Are model and production data reconciled on a defined cycle?
• Do the safety instrumented functions protecting the tie-back follow IEC 61511 lifecycle practice?

Conclusion

In a complex high-pressure tie-back, flow assurance is mostly won or lost on procedure discipline. Check that the operating strategy is documented and still matches the field's fluids. Test the injection system against API 17TR5 principles, and trend it continuously. Simulate transients before you run them, and make sure shutdown, restart and plug-response procedures are integrated across disciplines. For a first step, pick the highest-risk transient, usually a prolonged shutdown at current water cut, and walk through it with subsea, process and control room staff against the actual procedure. Where assumptions have gone stale, revise them before the next event forces the issue.