Arild Wilson, Carl Hartmann
Norsk Hydro a.s.
Oslo, Norway
The design of the tie-in of the Gamma North field's horizontal well to the Oseberg field complex required extensive analysis of the physical properties, hydrate evaluation, material studies, and multiphase flow.
The physical property generation was based on inhouse procedures that to a large extent are applications of the Gas Processors Association (GPA) developed techniques and method.
Experimental verifications have confirmed the need for these predictions.
OSEBERG DEVELOPMENT
The Oseberg field, located about 125 km (78 miles) west of Bergen, was discovered in 1979 (Fig. 1). The first well drilled revealed a large gas cap. However, the next well penetrated a substantial underlying oil zone. This turned Oseberg into one of the larger oil fields on the Norwegian shelf.
A field development plan was filed with the authorities late in 1983. Two phases of development were described:
Phase 1 would be on stream in 1988. It was to consist of a process and quarters platform (A) connected by a bridge to the drilling platform (B). This constitutes the field center (FC).
Phase 2 was planned to consist of a drilling, quarters, and water-injection platform. First stage separated oil and gas would be routed to the FC for further processing when capacity became available. The Phase 2 platform was planned to be on stream in 1996 but now plans are to have it on stream by 1991.
Selection of gas instead of water for pressure maintenance has, in addition to improved reservoir mapping, increased the recoverable reserve potential from the initial 153 million standard cu m (scm; 962 million bbl) of oil and 90 billion scm (3.2 tcf) of gas to the more recent values of 230 million scm (1,447 million bbl) of oil and approximately 92 billion scm (3.25 tcf) of gas.
This increase has resulted in a need to produce more reserves in the same time period. Hence, the Phase 2 platform is now being equipped with processing facilities to increase the field's total processing capacity.
The present FC capacity is approximately 300,000 bo/d. The Phase 2 platform will process 110,000 b/d.
A revised field development plan describing the new situation was filed in the autumn of 1987.
All associated gas will initially be reinjected into the Oseberg reservoirs; however, to provide the necessary gas volume required for pressure maintenance, two external gas sources are being developed: The Troll gas injection project, TOGI, and Gamma North.
After the oil production ends about year 2003, the Oseberg field will be a gas producer, and some 92 billion scm of gas will be produced.
GAMMA NORTH
The Gamma North reservoir is found some 9 km (5.6 miles) southwest of the Oseberg 2 platform location. This structure contains mainly a gas reservoir with a thin underlying oil column and an aquifer. The reservoir will be produced from a subsea well tied back to the platform with three lines: flow line, service line, and umbilical for hydroelectric well control.
The well will penetrate the reservoir horizontally in the thin oil column (Fig. 2). This will allow the simultaneous production of gas and oil. After the oil has been depleted, the well will be recompleted in the gas zone. The initial production will consist of gas, oil, and water.
SYSTEM CHARACTERISTICS
The reservoir pressure and temperature are relatively high, approximately 320 bar (4,626 psi) and 104 C. respectively. Also, the well fluid has a CO2 content leading to a high CO2 partial pressure. These facts in connection with the simultaneous production of oil, gas, and water through a 9-km flow line identified some significant design challenges. In short they can be summarized as:
- Nonequilibrium production of oil, gas, and water
- Special material requirements of the flow line
- Flow line dimensioning with respect to material, operation, multiphase flow, and turndown ratios
- Hydrate control
- Insulation requirements
- Operational models during normal production and planned or unplanned shutdowns of the line or well.
In addition, at the platform end the Gamma North production has to be choked down to the platform inlet separator pressure of approximately 70 bar (1,012 psi). This pressure drop would lead to multiphase supercritical choking and, in certain operating modes, severe temperature and metallurgical requirements.
A comprehensive design data study was initiated to establish all physical properties (Table 1) relevant for the design of the system. Where necessary, experimental verifications were performed.
FLUID CHARACTERIZATION
The reservoir gas and oil were characterized using the standard procedures by A. Wilson, et al.1 The fluids were described individually and were checked against experimental data. Then a combined characterization was performed. Because of uncertainties in production mechanisms in the horizontal part of the well, a swelling test was performed to simulate different production scenarios and to check the process simulators used .2 3
SYSTEM ENTHALPY
Due to the complex nature of the production mechanism from the reservoir, it was important to establish a correct enthalpy for the pipeline system to give the correct basis for insulation design.
The theoretical simulations were checked against an ingenious experiment which measured the enthalpy of a system which in its design modeled the pipeline conditions (Table 2).
MULTIPHASE CRITICAL CHOKING
The multiphase critical choke at the platform end was modeled theoretically. The model was inserted into the process simulator.
Crucial input data for the choke were speed of sound values for the system. These data were measured by the acoustic resonator developed through GPA project No. 831.
WAX SCALING AND FOAMING
Experiments were performed on the fluids to establish their characteristics with respect to wax and scaling. Due to the high system pressure, foaming was decided to be of minor importance for the flow line design (Fig. 3).
HYDRATE CONTROL
Hydrate predictions and hydrate distributions in the different fluid phases were predicted by the newly developed in-house technology.3 Experimental verification was performed to verify the theoretical model (Fig. 4).
PHYSICAL PROPERTIES
All relevant physical properties in addition to what is described above were prepared by the in-house simulation technology.
The results from the analysis indicated that:
- Gas composition was successfully established by characterization and gas blending.
- Simulated properties adequately match experimental data for: oil hydrate curve, methanol distribution, enthalpy difference, vapor-liquid equilibrium, and speed of sound.
- Simulated properties deviated from experimental figures for: interfacial tension and water saturation (high temperature).
- Emulsion and foaming were identified as potential problems.
IMPACT ON TOPSIDES
The subsea production system design was initiated after the platform topside was in the detailed engineering design phase, meaning that most of the boundary conditions on the platform topsides towards the subsea system were already more or less fixed. It was soon evident that the most important aspects on the platform topside were: expected slug sizes and frequency, inhibitor requirements, and temperature and capacity limitations during flow line shutdown activities.
To adequately map all these parameters, an iterative dialogue and design procedure between the topside design and subsea engineering design teams took place to enable the best design to be developed.
MULTIPHASE FLOW
All multiphase flow analyses were done based on the technology developed at the Tiller Laboratory in Trondheim, Norway.
The design data and physical properties were loaded into the Olga program. A comprehensive study of different operational scenarios was undertaken to establish the correct flow line design criteria and operational mode. The gas-oil ratio is expected to increase from 150 to 8,000 scm/scm during the field's producing life (Fig. 5).
FLOW LINE DIAMETER
The corrosive nature of the flow line fluid, combined with high temperature and pressure at the wellhead, soon ruled out carbon steel as pipeline material.
Duplex or flexible systems in plastic material were evaluated.
The final selection to a large degree was also dependent on system and installation price levels.
However, it was of utmost importance to ascertain a correct flow line diameter as early as possible. This was to allow the platform topside engineering team to continue design of the platform riser and topside piping system. A required diameter for a flexible system and one for a steel pipe was established.
HYDRATE INHIBITION
The hydrate prediction and inhibition studies soon revealed that hydrate prevention by continuous injection of inhibitor, i.e., methanol, would require volumes in excess of all practical limits for the platform storage and supply facility.
Therefore, the inhibition of the line had to be based on insulation. However, it soon became evident that during a shutdown of the flow line over an extended period of time, no viable insulation method would be sufficient.
An additional means of line hydrate prevention would therefore have to rely on pressure relief. Scenarios were now defined describing planned and unplanned shutdown and subsequent startup situations.
Also, each case was checked for temperature drop across the platform-located multiphase choke.
From these evaluations, the final requirement to flow line insulation could be established.
FINAL DESIGN
The Gamma North production system design has now been concluded. All the fabrication contracts for equipment, flow lines, service line, and well control have been initiated. The supporting platform systems are in the fabrication phase.
The production system consists of a subsea satellite well producing into a 9 km long flexible flow line of 8-in. OD. The line has basically two layers of insulation on the seafloor where it is covered by rock, and four layers in the exposed ends and riser areas. Hydrate prevention is further achieved by both methanol injection and pressure relief.
During normal operation inhibitor injection is not required.
No slug problems are foreseen during a wide range of system turn down ratios.
ACKNOWLEDGMENT
The authors wish to thank the partners in the Oseberg licenses who have supported the publication of this article. The Oseberg partners are: Statoil, Saga Petroleum a.s., Elf Aquitaine Norge A/S, Total Marine Norsk A/S, Mobil Exploration Norway Inc., and Norsk Hydro a.s.
Furthermore, we would like to express our recognition to Ali Majeed, Norsk Hydro a.s., for his excellent coordination and performance of the Gamma North experimental and theoretical work program.
We would also like to commend the experimental work by D.B. Robinson & Associates, Wiltec Research Co. Inc., and Sam Colgate, University of Florida, for performing parts of the experimental program. Our thanks go to O.A. Asbjornsen at the University of Maryland for his modeling efforts.
REFERENCES
- Maddox R.N., Erbar, J.H., and Wilson, A., "C6+ fractions affect phase behavior," OGJ, Aug. 21, 1978.
- Erbar, J.H., "MaxiSim - An Interactive Program for Process Simulation and Design," Stillwater, Okla.
- OverA, Sverre, and Majeed, Ali., "NEWS," Norsk Hydro a.s, Lysker, Norway.
Copyright 1990 Oil & Gas Journal. All Rights Reserved.