Case Study: Multiphase Flow Simulation Technology to Plan Pigging Programs

• More by Author
• Comments
• Send to a Friend
• Print

Over the years, the operation of multiphase subsea production systems has presented a variety of challenges to upstream operators. Multiphase pipeline simulation has made significant progress over the last two decades, and is now able to handle complex converging and diverging networks, and incorporate advanced thermodynamic calculations to facilitate modeling the phenomena observed in field operations. Today, real-time, online, transient multiphase flow simulation technology exists that allows the operator to perform flow assurance on a day-to-day basis. The goal here is to illustrate that not only are these simulators necessary, but they can simplify complex operational decisions to the push of a button.

Pigging Utility

In the case presented here, the operations team needed a utility (a software solution) from which its pigging campaigns could be planned. A pigging utility was developed that provides a way to plan these pigging campaigns to ensure that slug catchers will not flood during the process.
The pigging planner utility is built on top of an existing online transient multiphase pipeline model. This model also includes some dynamics of the slug catcher, including its level indications. The goals of the online model are different from that of a design tool; these differences are highlighted in Table 1.

Design Stage

The online model was built and tuned to match the steady-state total liquid holdup predicted by the design tool for the 58-mi, 36-in. subsea three-phase pipeline before site installation. After the online model was installed and tuned to field data, collected from various pigging campaigns, it was observed that the holdup was being under-predicted by as much as 33 percent. This necessitated changes to the pipeline model assumptions in order to replicate observed field behavior (Fig. 1).

One source of error was the bathymetry data. A smoothed version, instead of the actual bathymetry, was entered into the design simulations. This changed the angle-class distribution, which had a significant impact on the holdup. In addition, there were key fluid property differences between the field-tuned model and the design model, as summarized in Table 2.

HMI Interface

Planning of the pigging campaign begins with the current estimated pipeline state from the online model
(Fig. 2). Alternatively, several "condition" files can be generated that are states to which the pipeline will be brought before start of the pigging campaign. Any of these can be used as the starting point for campaign planning.  

Usability

The application was wrapped in a nice operator-friendly interface. Some of the salient features of the pigging utility that highlight this are:
- Avoiding usage of modeling terms such as boundary conditions, outlet pressure (instead use SC pressure).
- Outputing post-processed quantities that are relevant to operations. For example, indicated level of condensate is not the actual condensate level in the slug catcher because water swept out during pigging usually exceeds the limits of the water level indicator. Thus, the condensate level is condensate plus water.
- The operator is able to specify condensate and water loading instead of a full wellhead stream composition, which is unavailable.
- Ability to use different units sets. Although, in this application, the units are standard oilfield units, in some parts of the world, SI or some other units may be used. Users should always be able to input model parameters in units with which they are familiar.

Applications

Real-time pipeline monitoring using online simulators is the first step in leveraging the technology. Some of the applications include:
- Providing real-time estimates of liquid holdup, pressure, temperature and flow rate along the length of the pipeline
- Tracking pigs and pig-generated liquid slug volumes as they move
- Predicting closeness to hydrate formation, wax deposition, and asphaltene precipitation along the line.  

These applications also have look-ahead capabilities, i.e., to automatically forecast the state of the facility some hours into the future to help prevent bad events. For example, in a deepwater system, a cool-down look-ahead module can take the current state of the pipeline and estimate the time available before hydrates will form, if the production were to shut down immediately.  

Leak detection of subsea multiphase flow lines, especially in deep water, is feasible only using model-based methods. Studies have shown that model-based leak detection which combines statistical analysis of differences between model-predicted and measured quantities (such as inlet pressure) are the only reliable method for detecting leaks in transient multiphase production systems.

Inferential metering is another application of online multiphase flow simulation that could save millions of dollars in capital and operational expenditures. Inferential metering is software-based multiphase flow metering that utilizes pressure, temperature and choke measurements typically available down-hole and on the production tree.  This application replaces physical multiphase flow meters (MPFMs).  Using inferential metering, individual well flow rates are available in real-time, without the installation of costly MPFMs.

Conclusion

Online simulation tools can be created that are fit-for-purpose and customized to each field. With a field-tunable, online, multi-phase pipeline model as its backbone, a simple pigging planner utility was developed. This utility makes planning of complex pigging campaigns simply a matter of adjusting two or three inputs (launch time in this case) until the slug catcher level is within safe limits. Depending on the operational issues specific to a field, different utilities can be created that help ease the burden of operating that field.

• More by Author
• Comments
• Send to a Friend
• Print