Improved Prediction of Wet-Gas-Well Performance
- Pieter Oudeman (Koninkijke Shell E and P Laboratorium)
- Document ID
- Society of Petroleum Engineers
- SPE Production Engineering
- Publication Date
- August 1990
- Document Type
- Journal Paper
- 212 - 216
- 1990. Society of Petroleum Engineers
- 5.6.4 Drillstem/Well Testing, 5.3.2 Multiphase Flow, 1.6 Drilling Operations, 5.2.1 Phase Behavior and PVT Measurements, 5.6.8 Well Performance Monitoring, Inflow Performance, 1.14 Casing and Cementing, 2.2.2 Perforating, 4.6 Natural Gas, 3.1 Artificial Lift Systems, 3 Production and Well Operations
- 0 in the last 30 days
- 558 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 10.00|
|SPE Non-Member Price:||USD 30.00|
A series of field tests on liquid loading of gas wells demonstrated that various types of behavior could be encountered in such wells. On the basis of the test results, an improved model for predicting the performance of gas wells producing liquids was formulated. This model takes into account multiphase reservoir performance and the vertical flow performance of the tubings for wet gas. It provides improved estimates for performance of the tubings for wet gas. It provides improved estimates for gas deliverability and the remaining life of watering-out gas wells, compared with common methods that calculate "critical" conditions at a single point (usually the wellhead) in the tubing string.
An increasing number of gas wells drilled into the northwest European Permian basin have experienced water influx. This usually happens after 3 to 8 years of production when, owing to depletion, aquifer influx and (partial) flooding of the reservoir occurs. Water coning may precede water influx. Water production initially affects well productivity only marginally; at high gas rates, the well's capacity to produce liquids continuously to surface exceeds the rate of water influx. With continuing depletion, however, the gas rates decrease and liquids can accumulate in the wellbore. This liquid holdup exerts an additional pressure on the reservoir, further curtailing production. An increasing water saturation around the well-bore aggravates this effect. Eventually the well will "load up" with liquids and cease to flow. Either workover techniques for shutting off the water-producing zones (cementation and reperforation) to improve the well performance (recompletion with small-ID tubing) and to lower the performance (recompletion with small-ID tubing) and to lower the well-head pressure (compression) or artificial lift (e.g., plunger) may be required to restore production. These remedies for loading are costly and time-consuming. Hence, accurate methods of determining the optimum timing of workovers, installing compression, and evaluating their eventual results are essential to ensure that investments in watering-out gas fields are justified, that forecasted gas production meets negotiated contracts, and that additional wells are drilled in time. The following section briefly reviews published methods for predicting loading and their drawbacks. Then the proposed model, predicting loading and their drawbacks. Then the proposed model, which circumvents these drawbacks by taking into account both multiphase reservoir performance and tubing performance, is discussed. A comparison between field test data and model calculations demonstrates the validity of the approach.
Traditionally, "liquid loading of a wet gas well" was used to indicate that the bottomhole flowing pressure (BHFP) of the well could not be determined accurately from the surface production data and that liquid was produced to surface irregularly, so sampling to determine liquid (condensate and/or water) production gave erroneous results. These phenomena can be considered to be related to a change in flow regime from continuous (annular) mist flow to intermittent flow patterns, such as slug or churn flow. A large pressure drop over the tubing and rapid fluctuations in the gas/liquid pressure drop over the tubing and rapid fluctuations in the gas/liquid ratio are typical of the latter flow regimes. Of course, the change in flow regime does not necessarily mean that the well will cease to flow; this is determined by the total system performance, as shown by the following discussion of methods for predicting the onset of loading. An early approach to predicting the onset of liquid loading in gas wells was to define a minimum gas velocity at wellhead conditions of 5 ft/sec. Later work showed that wet-gas wells could cease flowing when wellhead velocities were less than 16 ft/sec. This indicates that the 5-ft/sec criterion cannot be considered generally applicable. This restriction also applies to the approach advocated by Ilobi and Ikoku that is based on a model for tubing-wall liquid film transport. Comparison of the model with available field data did not allow an unambiguous identification of the film flow conditions under which wells would load up. Hence, general use of this method cannot be recommended.
|File Size||325 KB||Number of Pages||5|