MULTI PHASE LOW CONCERNS GUIDE TOGI SYSTEM DESIGN

July 23, 1990
Morten N. Lingelem Henning Holm John Meling Norsk Hydro A.S. Oslo Plans to move raw gas condensate from the Troll field to the neighboring Oseberg complex for processing illustrate the design concerns for long-distance multiphase flow. A pipeline-monitoring system has been designed for the Troll-Oseberg gas-injection (TOGI) system in part to alert operators to hydrate formations in the system.
Morten N. Lingelem
Henning Holm
John Meling

Norsk Hydro A.S.
Oslo

Plans to move raw gas condensate from the Troll field to the neighboring Oseberg complex for processing illustrate the design concerns for long-distance multiphase flow.

A pipeline-monitoring system has been designed for the Troll-Oseberg gas-injection (TOGI) system in part to alert operators to hydrate formations in the system.

Transport of unprocessed gas condensate well fluids has been used for in-field flowlines for several years. Recently, the technology has been extended to transportation over long distances and with more complex fluids and in cold environments.

The requirement for detailed knowledge in the areas of multiphase flow and hydrate prevention has hence become increasingly more important.

Multiphase flow and hydrate-prevention technologies, therefore, influence the design of long-distance, gas condensate pipelines. And Norsk Hydro A.S.' TOGI development, for which an unprocessed gas condensate well fluid will be transported from the Troll field 48 km (30 miles) to the Oseberg field, is an example.

The TOGI system will move more than 10 million standard cu m/day (MMscmd) over to Oseberg for processing.

For multiphase transport lines starting on shore or at a manned offshore platform, it is possible to control liquid accumulation through pigging. In this way, production at low flowrates which promote liquid accumulation can be tolerated because regular (i.e., daily) pigging can be carried out to remove excessive liquid.

Multiphase transport lines originating subsea, however, normally cannot be pigged on a regular basis because the pigging operation requires use of a drilling rig. For such lines, the importance of detailed understanding of the multiphase flow is paramount, and determination of a proper operational envelope is one of the most important tasks of the early design.

The TOGI system, which relies on multiphase transport for more than 10 million standard cu m/day (MMscmd) over 48 km, will be used as an example.

THE DEVELOPMENT

Operator for the TOGI development is Norsk Hydro A.S. Participants in the Troll field are the Norwegian State (62.696%), Statoil (11.880%), Shell (8.288%), Norsk Hydro (7.688%), SAGA (4.080%), Elf (2.00%), Conoco (1.184%), Mobil (1.184%), and Total (1.00%).

The TOGI production system will transport gas from the Troll field through a 20-in. pipeline 48 km to the Oseberg field where the gas is to be processed, compressed, and injected into the gas cap of the Oseberg reservoir (Figs. 1 and 2).

The gas injection improves the total oil recovery by more than 7% over the originally planned water-injection system.

The TOGI subsea station is located in 303 m of water at the southwestern part of the Troll East field. There are five wells remotely controlled from Oseberg by an electro hydraulic control system.

Methanol is pressurized at Oseberg and injected through a service line into the wells at Troll.

The TOGI fluid is at its hydrocarbon and water dew points at reservoir conditions. Maximum liquid drop out is around 5% (mass). No liquids are produced from the reservoir.

LIQUID ACCUMULATION

Liquid accumulations are not necessarily a problem as long as they stay where they are. The problem is that they tend to move.

Fig. 3 shows a typical curve in which total liquid volume is plotted against the flowrate. If prolonged operation at a low flowrate is followed by a higher flowrate, the difference in equilibrium liquid volume between the two flowrates will be disposed of as indicated.

The liquid will be disposed of over a relatively short time frame. There are generally two ways to handle it: "direct processing" with limited buffering, and buffering in a slug catcher for later processing.

For pipelines transporting gas condensate, direct processing is rarely practical because the variation in flowrates of liquids is several orders of magnitude between steady state and when gross amounts of liquids are disposed of. Hence, a buffer tank is required. Normally the inlet separator is designed to handle this task and is known as a slug catcher.

For multiphase lines terminating onshore, it is economical to provide slug catchers of significant dimensions, possibly as caverns. For multiphase pipelines terminating at offshore platforms, however, space limitations severely restrict the size of any practical slug catcher, and hence knowledge of liquid accumulations and flow dynamics of the pipeline becomes essential.

There are three main causes for liquid accumulation: liquid inherently condensing from the fluid, low flowrates causing inadequate liquid transport, and large inclinations limiting liquid transport.

The importance of detailed knowledge of the pipeline elevation profile cannot be too strongly emphasized.

Fig. 4 shows typical curves for liquid vs. flowrate for different pipe inclinations. For a given inclination there is a quite distinct flowrate below which liquid holdup increases dramatically.

The TOGI pipeline crosses one of the steepest areas of the North Sea, the western slope of the Norwegian trench. The average inclination, however, rarely exceeds 1.

Locally there are surface irregularities. At Troll there are nearly circular depressions known as pock marks, and running north-south on the western slope of the Norwegian trench there are narrow surface depressions (Figs. 5, 6, and 7).

Locally, these surface depressions give pipeline inclinations as high as 5 for the route chosen.

Areas with even higher inclinations were avoided during selection of the pipeline route.

The TOGI pipeline has been found to have a minimum flowrate of around 5 MMscmd above which operation is acceptable with respect to liquid accumulations.

OPERATIONAL ENVELOPE

The operation is limited by pressure loss at high flowrates and excessive liquid accumulations at low flowrates.

The pipeline design must hence be a balance between the two.

It is evident that for a given minimum inlet pressure and slug-catcher size, the longer the pipeline the lower the flexibility of the operation, as depicted in Fig. 8.

The TOGI development benefits from the vast volumes of gas in the Troll reservoir and resulting very moderate decline in pressure over the 18-year design life.

With a minimum flowrate of around 5 MMscmd and a maximum flowrate in excess of 10 MMscmd, the TOGI line has an acceptable operational flexibility.

The liquid outflow resulting from an increase in flowrate may appear at the pipeline outlet as a slug (holdup close to 1) or be more spread out.

The formation of slugs may be more severe both in terms of transients in the process facilities and the mechanical loads on piping and supports.

During design of TOGI, line simulations have shown that slugging is a smaller problem than was anticipated.

It was found that the interaction between liquid slugs accumulating at the riser base and the compressor train was such that the compressor was capable of sucking out liquid very quickly after slugs of any significant size formed and blocked the flow of gas. Hence no severe liquid accumulations were allowed to build up at the riser base.

It was also found that slugs formed in the rough terrain in the western slope of the Norwegian trench were quickly dissipating in the smooth slightly upwards inclined part of the pipeline before reaching the plateau around Oseberg.

For a pipeline which is inclined predominantly downwards with the pipe crossing a number of surface irregularities, it is conceivable that slugs will not dissipate but rather merge to form larger slugs causing very substantial liquid surges and transients in the downstream process equipment.

HYDRATE FORMATION

Hydrate formation is possible for all gas-condensate fluids operating at typical pressures and for temperatures found in the North Sea and in more northerly waters. The hydrate-formation curve for the TOGI fluid is given in Fig. 9. The TOGI fluid hydrate formation curve differs little from typical gas-condensate systems.

In order to inhibit the hydrate formation, methanol or glycols may be injected. Other inhibitors are in use but have limited popularity compared to methanol and glycols.

The TOGI system uses methanol rather than glycols, one main reason being the lower viscosity of methanol.

The consumption of inhibitors will normally be very close to the amount of water produced. Injected methanol ending up in the water phase ( 60% mass) is regenerated and reused.

For most gas-condensate systems as in the case for TOGI, the water produced is only what condenses from the gas.

A key value is the amount of water in saturation. TOGI has carried out an extensive effort to determine this value as well as the distribution of methanol between the phases.

With a typical injection of between 30 and 35 cu m/day, the potential for savings through economizing methanol injection is substantial.

Formation of hydrate blockages is hazardous because they may break loose and damage the pipes and pipeline.

Furthermore, hydrate blockages are not easily removed.

In order to decompose hydrates, the pipeline must be depressurized to obtain a reasonable heat transfer causing decomposition of the hydrates.

An often-overlooked consequence of gas-pipeline ruptures is the potential formation of hydrates as water flows into the pipeline. For pipelines in waters deeper than 300 m (984 ft), hydrates will form at typical temperatures found in the North Sea.

Decomposition of such hydrate blockages may be difficult and may cause the repair operation to be more time consuming than expected.

TOGI MONITORING, LESSONS

Monitoring of gas pipelines is desirable both in terms of safety and operational flexibility.

Monitoring systems for single-phase pipelines are commercially available. For multi phase lines, however, the market for monitoring systems is meager.

For the TOGI multiphase pipeline, it was decided early in the development to include a monitoring system. The monitoring system's main functions are liquid outflow forecasts, leak detection, hydrate-blockage detection, and data acquisition.

The monitoring system uses a number of measurements (Fig. 10) to serve as boundary conditions for a numerical model of the pipeline and downstream process equipment.

The model used is based on the OLGA multiphase flow code developed by IFE/Sintef.

The monitoring system has two main processes (Fig. 11). The first is a real time part which handles leakage and hydrate-blockage detection algorithms as well as general housekeeping functions.

The second process runs around 40 times real time and performs forecast calculations for liquid outflow.

The monitoring system consists of two Hewlett-Packard 835-SRX computers (Fig. 12).

Normally, one computer runs the real time process while the other carries out the forecast calculations.

If one of the two computers fails, an autodetection system starts both processes on the remaining computer. This reconfigured system can carry out all required functions with an acceptable loss in performance for the forecast calculations.

The monitoring system will be operated from the Oseberg field center central control room.

Through the TOGI project, long-distance transport of an unprocessed gas-condensate well fluid has been shown to be an attractive and viable alternative to processing at the location of production.

For future application of the TOGI technology to other developments, the following points should be taken into consideration:

  • Liquid drop out. The amount of liquid condensing out for a wide range of pressures and temperatures must be known.

    Water condensation is essential for determination of a proper amount of hydrate-inhibiting chemicals.

  • Pipeline elevation profile. The pipeline elevation has a profound influence on the pipeline's properties to accumulate liquids. At low flowrate, the effects of local surface irregularities, when added together, may be the main cause of liquid accumulation.

    In many instances, depth information for every 0.10 m or better may be called for in order to have a representative elevation profile.

  • Operational envelope. It is important to establish the minimum as well as maximum flowrates and ensure a comfortable range of flowrates is available for pipeline operation.

    Also, the dynamic properties of the pipeline need to be established in order to determine the amount of liquid disposed of during changes in flowrate or during normal operation at low flowrates in order to size the downstream slug catcher and downstream process equipment.

    For systems not having desirable properties in terms of liquid outflow, regular pigging may prove to be economical for solving such problems. For a number of scenarios, regular pigging may be a less expensive solution than subsea pumping/compression, for example.

  • Hydrate prevention. Prevention of hydrate formation is an important part of any transport of unprocessed well fluid.

    Injection of hydrate inhibitors is expensive both in terms of the injection system and chemicals used. There are strong incentives to optimize the chemical injection.

    Research with the objective of finding inhibitors that inhibit the formation of blockages rather than the formation of crystals should be encouraged.

    For deepwater submarine pipelines, a rupture may lead to massive hydrate formation once water is flowing into the pipeline.

  • Pressure development in the producing reservoir. For most reservoirs, the pressure declines appreciably as the gas is produced. A multiphase pipeline, and in particular one originating at a subsea installation, will normally not have any compression facilities available.

    The design must therefore be a compromise between operational flexibility when pressure levels are high (low velocities) and capacity when pressure levels are low (high velocities). There will be a trade off between installation of additional pipelines and installation of compression equipment (subsea or surface).

For most fields there will be a net profit in use of unprocessed multiphase transport for the first stages of development because investments may in this way be delayed and additional investments made at a time when knowledge of the field, in particular reservoir structure and behavior, has been significantly enhanced.

ACKNOWLEDGMENT

The authors express their appreciation to Norsk Hydro for permission to publish this article.

Copyright 1990 Oil & Gas Journal. All Rights Reserved.