Read Production/Facilities Overview text version

OVERVIEW

Production/Facilities

Innovation is defined as the introduction of a new or different idea, method, or device. Continuous production innovation is the source of change that is essential to your project's success in a changing environment. Innovation is central to project-value creation and successful project completion through cost and schedule reduction. The success of offshore production/facilities projects during the past 4 decades can be attributed largely to technological innovations, hard work, creativity, and "out-of-the-box" thinking. Techniques have evolved since 1859, from collecting crude oil in wooden barrels from open gushers relying on natural pressure and rudimentary pumps [that recovered less that 10% of the original oil in place (OOIP)], to complex gas-, thermal-, and chemical-injection techniques producing crude oil and gas from remote distances relying on artificial lift recovering as much as 50% of the OOIP without losing a drop and 75% of the gas without a leak.

Dino

Additional Production/ Facilities T echnical Papers

Available at the SPE e-Library:

www.spe.org

· SPE 80507 Transporting an Oil and Gas Mixture in a Gathering System at White Tiger Oil Field · SPE 84226 A Mechanistic Heat-Transfer Model for Vertical Two-Phase Flow · SPE 84346 Liquid Management on Canyon Express

Improved production-recovery technologies have greatly reduced project costs and schedules as well as environmental effects. Notable production/facilities technology innovations during these 4 decades include improved produced-water treatment; improved sour-gas sweetening and treating of gas and liquids; increased efficiency of artificial-lift systems; production of coalbed methane; development of economical gas-to-liquids conversion processes; improved design of glycol-dehydration systems; incorporation of advanced data management and real-time optimization; invention of novel leak-detection and -measurement systems; improved multiple designs for offshore platforms and decks; emerging multiphase-fluid transfer, slug handling, and downhole fluid-separation technologies; steadily improved multiphase modeling and deepwater liquid-management techniques; improved vapor-recovery techniques; and implementation of greenhouse gas, safety, and environmental-management programs. Do you believe in innovation on your project? It is the source of your future project cost and schedule reductions. Everyone must try to be innovative; the challenge is to discovJPT er how to unleash your inner innovative self! Galen Dino, SPE, is Project Manager in the Project Services Group of Paragon Engineering Services. He had 23 years' experience with several operating and service companies before joining Paragon in 2000. Dino holds a BS degree in chemical engineering from Louisiana State U., with advanced graduate studies at the U. of Delaware. His main interests are project management, oil and gas separation, gas processing, and multiphase-fluid flow, and he has authored two papers on project innovation. Dino is a Review Chairperson for SPEPF, serves on the JPT Editorial Committee, is on the SPE Facilities and Construction Advisory Committee, and is Chairperson of the SPE Production Systems and Facilities TIG. He is a Registered Professional Engineer.

Available at the OTC Library:

www.otcnet.org

· OTC 15186 Impacts of Transient Analysis on Kuito Production Operations · OTC 15306 Deepwater Pipeline Spanning and Rectification: Lessons Learned

28

DECEMBER 2003

P r o d u c t i o n / Fa c i l i t i e s

Natural-Gas Vapor Recovery Using Nonmechanical Technology

There are approximately 250,000 tank batteries in the U.S. Tank vent gas results from flash, standing, and working losses. Between 8,000 and 10,000 mechanical vapor-recovery units (VRUs) are installed at oil production sites, with four tanks typically connected to each VRU. Most of these tanks are fixed-roof storage tanks. This project demonstrated the use of an alternative method for handling low-pressure natural gas that typically is vented to the atmosphere, flared, or recovered with a mechanical compressor at oil and gas production facilities. This application uses a venturi jet ejector to compress natural-gas vapors from crude-oil and water storage tanks to an intermediate pressure for use at the location or, ultimately, delivery into the sales pipeline. Theory According to Bernoulli's equation, if no work is done on or by a flowing frictionless fluid, its energy from pressure and velocity remains constant at all points along the streamline. As a result, an increase in velocity is always accompanied by a decrease in pressure. This principle can be used to collect a low-pressure natural-gas stream with a high-pressure motive-gas stream for entrainment and compression to an intermediate pressure. A venturi jet ejector is a mixing and pressure-increasing device consisting of a nozzle and venturi. As Fig. 1 shows, the nozzle receives the motive fluid (e.g., natural gas) from a high-pressure source. As the motive fluid passes through the jet, velocity increases and pressure decreases. The increased velocity, plus the decreased pressure, causes suction around the nozzle. Low pressure gas around the nozzle is drawn into and mixed with the motive stream. The venturi is a section of pipe having a narrow diameter at the throat that widens at its terminal end. This increased diameter at the terminal end causes the mixed-fluids velocity to decrease and the pressure to increase. The venturi converts the high-velocity jet stream to an intermediate-pressure stream

This article, written by Technology Editor Dennis Denney, contains highlights of paper SPE 80599, "Vapor Recovery of Natural Gas Using Nonmechanical Technology," by Mark A. Goodyear, COMM Engineering; Alexandra L. Graham and John B. Stoner, SPE, Total S.A.; and Brian E. Boyer, SPE, and Lyle P Zeringue, . COMM Engineering, prepared for the 2003 SPE/EPA/DOE Exploration and Production Environmental Conference, San Antonio, Texas, 10­12 March.

for delivery to a system for this intermediate pressure. Equipment and Processes The El Ebanito oil and gas production facility is in Starr County, Texas, approximately 30 miles northwest of McAllen. The facility had an existing mechanical VRU that collected stock-tank gas, increased its pressure, and delivered the

Fig. 1--Venturi jet ejector. PI=pressure indicator, TI=temperature indicator

DECEMBER 2003

29

gas to the on-site booster compressor for sales to a gas pipeline. The Texas Commission on Environmental Quality and the U.S. Environmental Protection Agency require the facility to operate a vapor-recovery system on storage tanks. Initially, an electrically driven vane compressor was used to handle the vapors. Because of reliability issues, the unit was replaced with a mechanical VRU that used a natural-gas engine as the driver. It was estimated that the mechanical VRU operated less than 90% of the time. Because of this reliability level and resulting operation and maintenance cost, new options for vapor recovery were researched and a venturi jet ejector was selected. This project collects gas from vents, storage tanks, and pressure vessels. Venturi jet ejectors use the kinetic energy in a high-pressure motive gas to create a vacuum that can entrain and mix another source-gas stream. The mixed motive and source gases can be delivered to the suction of a compressor, a low-pressure separator, the fuel-gas system, or the flare. Thus, the unit is a closed-loop system that reduces or eliminates vent-gas emissions.

Information required for the venturijet-ejector design includes the following: · Operating pressure of low-pressure system (e.g., suction pressure of each stage of the on-site booster compressor). · Sources of high-pressure gas (e.g., compressor discharge, high-pressure gas to and from the glycol dehydration-unit contact tower). · Operating pressure of high-pressure gas system. · Current and future estimated volume of gas that could be compressed by the booster compressor (needed to ensure that adequate horsepower is available to compress intermediate-pressure gas from the venturi jet ejector). · Amount of spare horsepower available at the booster compressor for compressing intermediate-pressure gas from the venturi jet ejector. · Tank dimensions, operating pressure, location, and dimensions of vent piping. · Plot plan with location of existing tanks, compressors, and piping. · Volume of gas recovered by existing mechanical vapor recovery. Facility storage consisted of five fixedroof storage tanks (400-bbl capacity each)

and two gun-barrel tanks (750-bbl capacity each). The pressure-relief valve settings for the tanks were 0.3 psig of positive pressure and 0.3 psig of vacuum. The facility was configured with the vent from each tank routed to a common 6-in.diameter header pipe before collection at the mechanical VRU. The mechanical VRU increased the vent-gas pressure for delivery into the first stage of the existing on-site booster compressor. The on-site booster compressor received gas from other on-site low-pressure separators for compression before delivery to the sales pipeline. The flow rate of gas vented by the tanks was measured by use of an ultrasonic meter. Flow-rate readings were taken at 10-second intervals throughout a 24-hour period. This sampling period provided a complete 24-hour-period profile of the gas flow. While measuring the flow rate, a sample of the vent gas was collected. This sample was chemically analyzed for methane through decanes-plus components at a commercial laboratory. During the flow-rate measurement, the oil production was approximately 700 BOPD of 62°API oil. The pressure at

Resources

Echometer Well Analyzer

Data acquisition and analysis are performed using a powerful notebook computer housed with an accurate analog to digital converter in a rugged briefcase. A cost advantage is realized by using one electronics package with the desired sensors to obtain a complete well analysis. · · · · Liquid Level Instrument Dynamometer Pressure Transient Tester Motor Power Sensor

Echometer Company 5001 Ditto Lane Wichita Falls, TX 76302 Phone: (940) 767-4334 Fax: (940) 723-7507 e-mail: [email protected] www.echometer.com

30

DECEMBER 2003

the separator, upstream of the storage tanks, was approximately 80 psig. The measured 24-hour flow rate was 202,000 scf with an average flow rate of 140.3 scf/min. The flow ranged from 90 to 160 scf/min. The laboratory analysis of the vent gas yielded an energy content of 1,850 BTU/scf. To ensure that adequate capacity was available for future increased production, two venturi jet ejectors in parallel were proposed for the application. Venturi Jet Ejector 1 was designed to collect 200 Mscf/D, and Venturi Jet Ejector 2 was designed to collect 100 Mscf/D, for a combined design capacity of 300 Mscf/D. The unit was designed to maintain tank pressures in the range of 0.10 to 0.30 psig. Ejector 1 was designed to maintain tank pressures from 0.10 to 0.20 psig; Ejector 2 was operated when tank pressures exceeded 0.20 psig. Both ejectors deliver vent gas to the first stage of the on-site booster compressor for delivery to the sales pipeline. The source of motive gas for the venturi jet ejectors was gas from the on-site glycol dehydrator's contact tower. The motive gas to the venturi jet ejectors was con-

trolled with a pressure regulator and flow controller to maintain a design pressure of 850 psig. The discharge pressure from the venturi jet ejectors was approximately 25 psig. The pressure and vacuum setting for the tanks' pressure-relief devices are the same as the original settings when the mechanical VRU was operational. To ensure that positive pressure is maintained in each tank, the programmablelogic controller was set to maintain a positive pressure of 0.1 psig. The new system is designed to shut off the motive gas and shut down the unit if the pressure exceeds this set point. To ensure that air does not enter the system, an oxygen sensor was installed upstream of the venturi jet ejectors. The oxygen sensor shuts the system down if oxygen concentration in the vent gas exceeds 2.5 vol%. Results The unit began operation on 15 May 2002. Since startup, the system has collected approximately 175 Mscf/D of 1,850-BTU gas for delivery to the sales pipeline. At that recovery rate, the

amount of gas recovered is valued at U.S. $336,780/year, based on a gas price of U.S. $2.85/million BTU. At the time this paper was written, the unit had not experienced any downtime. Conclusions This project demonstrated that venturi jet ejectors that use high-pressure natural gas as a motive gas are a viable technology for recovering storage-tank vent gases. Because the venturi jet ejector is a pressure-increasing device, it can be used to recompress lowpressure gas from a variety of sources for delivery to the system (e.g., gas lift, fuel gas, or sales gas). Besides storage tanks, the technology can be used to recover vent gas from glycol dehydrators, heater treaters, and lowpressure separators. The concern about greenhouse gas emissions, ozone formation, and the need to conserve natural resources will require the recovery of more vent gas for JPT useful purposes.

For a limited time, the full-length paper is available free to SPE members at www.spe.org/jpt. The paper has not been peer reviewed.

DECEMBER 2003

31

P r o d u c t i o n / Fa c i l i t i e s

Application of Multiphase Pumps in a Remote Abu Dhabi Oilfield

A multiphase pump was commissioned recently in a satellite oil field onshore Abu Dhabi. The pump is used to boost the flow from a well in a remote desert location to a crude-oil gathering center 16 miles away. The full-length paper describes the multiphase-pump installations and the unique features of these facilities. Pump specifications and key design considerations are discussed. Introduction Huwaila field is an undeveloped oil reservoir in a remote desert location in southern Abu Dhabi. The field was discovered in 1965. The closest production facilities to Huwaila are in a gathering center in the nearby Bu Hasa field. The gathering center, called RDS-3, is 16 miles north of Huwaila field. The terrain is hilly with undulating sand dunes. Access is only by four-wheeldrive vehicles. Production from the field began using one well, Hu-44. Initially, the well was produced by natural flow to the surface and through the 16mile, 8-in. flowline to RDS-3. Since 1996, an electrical submersible pump (ESP) was used to produce the well. Well production during the first years of ESP use was characterized by rapidly increasing water cut. Following a single-well reservoir-simulation study, Well Hu-44 was sidetracked as an 1,800-ft horizontal well. Various options to flow the well to RDS-3 were evaluated. A project was approved to try multiphase-pumping technology. Both dynamic and positive-displacement pumps were considered. A discussion on the merits of each type and the gas volume fraction (GVF) are given in Appendix 1 and Appendix 2, respectively, of the full-length paper. On the basis of the actual flow rates and GVF at Huwaila, it became clear that the process parameters were more suited to positive-displacement pumps than to dynamic pumps. Vendor Scope of Supply The multiphase pump, together with associated electrical and control systems, was tendered as one package to ensure single-point responsibility and to avoid interface problems between different vendors. In developing the technical requirements of the tender documents, several vendors and end users were approached for their experience. There are no industry standards to cover design and manufacture of multiphase pumps. For twin-screw pumps, the design is usually in compliance with American Petroleum Inst. (API) 676, which deals with positive-displacement liquid pumps Because the pump vendor was German, it was agreed in the preaward meeting that Duetsche Industrie Norm (DIN) standards would be adopted for some components instead of API standards. For example, the load calculation of the timing gears was in accordance with DIN 3990 and not American Gear Manufacturers' Assn. standard 6010 as stipulated in API 676. Also, the performance-test procedures were not in accordance with API 675. Performance Characteristics of Twin-Screw Pumps The performance characteristics of the multiphase twin-screw pump are best discussed by examining its performance curve. The performance of a twin-screw pump is usually plotted as a curve between the flow and the differential pressure as shown in Fig. 1. Theoretical flow rate Qth depends only on the physical dimensions of the pump and the pump speed. Actual flow rate is less than the theoretical flow rate because of the fluid that leaks through the internal clearances of the pump. This leakage is often referred to as slip. Volumetric efficiency of the pump is the actual flow rate divided by the theoretical flow rate. Power required by a twinscrew pump is equal to the theoretical flow rate times the differential pressure across the pump minus power lost to friction. Internal clearances, differential pressure, liquid viscosity, GVF and screw geometry influence , slip. The size of the screws and the screw pitch determine flow losses. Twin-screw pumps require a minimum amount of liquid for lubrication, cooling, and sealing between the screws and liner. A patented feature of the pump selected for Huwaila is the internal circulation of liquid through the throttle bushing.

This article, written by Assistant Technology Editor Karen Bybee, contains highlights of paper SPE 81504, "Application of Multiphase Pumps in a Remote Oil Field Onshore Abu Dhabi," by Hisham Saadawi, SPE, and Sharif Al Olama, Abu Dhabi Co. for Onshore Oil Operations, prepared for the 2003 Middle East Oil Show, Bahrain, 9­12 June.

Special Design Considerations One project objective was to test and gain experience with multiphase-pumping technology in the Abu Dhabi operating environment. The choice of Well Hu-44 offered some unique challenges. Low-Viscosity Crude. The viscosity of Huwaila crude oil is 0.55 cp. The lower the viscosity, the greater the slip. There was no record of screw pumps used to handle such a low viscosity. A low viscosity means more power consumption by the twin-screw pumps for the same throughput because of the high slip. A conservative value for the volumetric efficiency had to be adopted because of the high slip. High Differential Pressure. The differential pressure needed at the design flow rate is 930 psi. This is considered too high for a single multiphase pump and is outside the experience envelope of the vendor. To overcome this uncertainty, two pumps were used in series, so that each pump can be operated within its operating envelope. However, this created another challenge. There is no industry experience with two multiphase twin-screw pumps operating in series. The control-system design had to take this into consideration. Because there will be a time lag between any variation in suction pressure and the response of the second pump, it was important to size the medium pressure header properly because the header will act as a surge vessel to absorb any pressure fluctuations until the control system of the second pump responds. A large header also would

32

DECEMBER 2003

cushion the effect of changes in GVF The . medium-pressure-header size was calculated to be 24 in. Corrosive Well Fluids. The well fluids are very corrosive. The gas contains 0.7% H2S and 7.0% CO2, and the formation water contains chlorides. Also, the presence of H2S meant that a simple mechanical seal, such as API plan 52, could not be used. Instead, the smart-seal system S3 was used. Harsh Desert Environment. The oil field is located in a hostile desert environment. Maximum ambient temperature during summer is 122°F The metal temperature in . the sun can be as high as 185°F The pro. grammable-logic controller and variable-frequency controls were installed in an air-conditioned container. The air-conditioning system is of the split type, with one unit running and one standby. Voltage Dips. Power is supplied to Well Hu-44 by a 33-kV electrical overhead line. The same overhead line supplies power to the electrical submersible lines in Bu Hasa field. The line can be subjected to voltage dips as high as 20%. This would cause the multiphase pump to trip, and an operation staff would have to drive to Well Hu-44. To overcome this problem, the uninterruptible power supply (UPS) was specified to be 240 volts instead of the conventional 24 volts. Thus, the control supply for the variable-frequency drive is fed from the UPS. Drain Facilities. There is no infrastructure at the well location. The provision of a blowdown or a closed drain system would have required a purge and/or a flare system. This would have meant additional costs and more complex operations. Instead, a drain vessel is provided that is positively isolated from the system. Quality Control The company appointed a third-party inspector (TPI) in Germany to monitor the quality-control aspects of the project. Shortly after contract award, a premanufacturing meeting was held at the vendor's works near Hanover, Germany. The meeting was attended by the company and the TPI. At the premanufacturing meeting, the inspection and test plan were finalized. Points of witness and hold were identified. The TPI reviewed all documents and related material certificates for the pump components, suborders, and welder's qualifications.

Shop Testing The pump package was subject to a comprehensive testing program in the vendor shops. The pump casing was subject to a hydrostatic test in accordance with API 675 to 1.5 times the design pressure for 30 minutes with no leakage. P e rf o r m a n c e testing was performed in accorFig. 1--Performance characteristics of a twin-screw pump. dance with the German standards for positive-displacement-pump testing, the shutdown valves was filled. The pumps Verein Deutscher Maschinenbauanstalten were first run with water before introduc(VDMA) 24284. The company accepted the ing well fluid. During the commissioning use of VDMA criteria instead of the API 676 period, various startup problems were overbecause of the uncertainty concerning the come. At the end of the commissioning slip. The test was performed using the shop phase, a site-acceptance performance test driver and water as the testing medium. was performed. The pump was run for 7 During the test, flow, pressure, temperature, days while all operating parameters were and power measurements were taken at five recorded. Pump performance was in accordifferent operating points. These points were dance with the specifications. achieved by varying the differential pressure across the pump from zero to a maximum in Conclusions 20% increments. The performance curve for The multiphase-pumping facilities onshore the pump was plotted for water and then Abu Dhabi were commissioned successfully during September 2000 and are currently corrected to an actual performance curve. The pump was run for 1 hour at the rated operational. Although it is too early to evalspeed with a 145-psi differential pressure. uate system performance fully, initial results The pump was run with air for 30 minutes have been very encouraging. Multiphase with a pressure ratio of 6. The maximum pumping can be a viable technological and cost-effective solution in developing small casing temperature was less than 176°F . The string test was the first time the vari- satellite oil fields in the Middle East. Two twin-screw pumps have been opous components of the multiphase pump were put together. The purpose of the string erated successfully in series, making it postest was to ensure the mechanical integrity sible to develop a differential pressure of the complete pump train and associated greater than what would have been equipment. Considering the field location, achieved by one pump. Multiphase twinit was more efficient to debug the system in screw pumps have been used successfully in pumping crude oil with a viscosity as the vendor shop rather than on site. low as 0.55 cp. Expanding the multiphase-pump-vendor Commissioning Before commissioning the system, a training scope of supply to include the electrical, course was run on site by the pump vendor control, and piping systems ensured that for operations and maintenance personnel. there was a single point of responsibility. The course covered classroom lectures as This facilitated project coordination and JPT ensured smooth commissioning. well as hands-on practical sessions. Calibration of field instruments and testing of the control loops were performed as For a limited time, the full-length paper part of the precommissioning procedures. is available free to SPE members at The cooling system was filled with water www.spe.org/jpt. The paper has not and glycol. The combined seal and lubricabeen peer reviewed. tion-oil system was filled with oil. Similarly, the hydraulic-oil cylinder for actuation of

DECEMBER 2003

33

P r o d u c t i o n / Fa c i l i t i e s

Promoting Real-Time Optimization of Hydrocarbon-Producing Systems

The term "real-time optimization" (RTO) has found its way into common usage in the oil and gas industry. However, RTO usually is used more as a slogan rather than describing a system or process that truly optimizes anything, let alone does so in real time. The full-length paper describes what RTO means in the exploitation of hydrocarbons and what technologies are available now and are likely to be available in the future. The paper also describes a new Society of Petroleum Engineers (SPE) technical interest group (TIG). motives may be incompatible and may not be achieved simultaneously, but these are reallife challenges. The objective function for a hydrocarbon-production system may be expressed as the finite sum of discounted cash flows during the project horizon. In practice, the optimal solution is that a model is assumed that determines the cash flows in time for certain assumptions and ultimately finds a maximum (or minimum) value for NPV while honoring system constraints. The numbered boxes in Fig. 1 represent a typical sequence of steps in the traditional optimization process of hydrocarbon-producing systems. While each of the steps is individually practiced today, there may be little coordination of these steps. For example, individual steps or parts of steps may be disconnected or bypassed. Orchestrating these steps will contribute greatly toward the practical realization of continuous (realtime) enterprisewide optimization.

This article, written by Assistant Technology Editor Karen Bybee, contains highlights of paper SPE 83978, "Promoting Real-Time Optimization of Hydrocarbon Producing Systems," by L.A. Saputelli, SPE, U. of Houston; S. Mochizuki, ExxonMobil; L. Hutchins, SPE, BP; R. Cramer, SPE, Shell; M.B. Anderson, SPE, and J.B. Mueller, SPE, Schlumberger; A. Escorcia, SPE, Halliburton; A.L. Harms, SPE, ConocoPhillips; C.D. Sisk, SPE, BP; S. Pennebaker, SPE, Baker Hughes; J.T. Han, SPE, Welldynamics; A. Brown, SPE, EPS; C.S. Kabir, SPE, ChevronTexaco; R.D. Reese, SPE, Case Services; G.J. Núñez, SPE, ITSVE; K.M. Landgren, SPE, Schlumberger; and C.J. McKie, SPE, and C. Airlie, EPS, prepared for the 2003 Offshore Europe, Aberdeen, 2­5 September.

Concepts and Definitions Optimization. Examples of what usually is meant by optimization in the oil and gas industry include the following. · Maximizing hydrocarbon production or recovery. · Finding the best solution subject to physical and financial constraints to produce a decision. · Maximizing net present value (NPV) through changes in capital expenditure Real Time. Real time includes measure- those decisions, concluding in new mea(CAPEX) and/or operational expenses ments, the subsequent analysis of those surements, thereby starting a new cycle. measurements, the evaluation of that analy- Fig. 2 presents this concept. (OPEX). There are different time scales of real time The term optimization usually is used sis, the decision-making processes that very loosely, whereas it should be defined result from the evaluation, and the conse- for different decision-making processes that rigorously and mathematically, while honor- quential changes in system response result- might more easily be thought of as being ing the real-life physical system constraints ing in actions carried out on the basis of time constants for the various processes. For example, the time conthat exist in the overall prostant may be on the order of duction process. seconds for fluid flow, minThe exploitation and proutes for separation, months duction of hydrocarbons for equipment wear, and involves a complex interacyears for reservoir exploitation of many variables from tion. When talking about reservoir and wells (subsuroptimizing the entire proface) to product conditioning duction system, the differand transport (surface). In ent time constants of the this system, optimization can system must be considered realistically be discussed only when moving from reserwithin the concept of a mathvoir to well, to surface facilematical process. This should ities, to export, to sales, and be thought of as maximizing to business headquarters. profit where profit is carefulThe different dimensionalily defined for the specific ty and discrepancies in problem. Profit could refer to scales represent tremendous NPV of the entire life of the challenges in performing field or net income derivable integrated decision making over the next 3 months. Fig. 1--Traditional hydrocarbon-producing-systems optimization. over time. Clearly, these two profit

34

DECEMBER 2003

Fig. 2--RTO.

Fig. 3--Optimization cycle.

Putting together the important factors relating to optimization and real time, perhaps the best definition of RTO is based on outcome, and the authors propose the following definition. RTO is a process of measure/calculate/control cycle at a frequency that maintains the system's optimal operating conditions within the time-constant constraints of the system. What can be optimized today should be optimized today, and it should be recognized that this will change in the future and that assumptions will need to be adjusted. Opportunities and Benefits Any technology or methods applied should provide business benefits, such as improved financial performance, reduced risk, increased reserves, improved manageability, or reduced environmental impact. RTO can deliver benefits in each of these categories, and extensive case studies are available in the literature. Business benefits include sustainable production improvements, improved well and facilities downtime, and substantial OPEX and CAPEX reductions. Further benefits are safety improvements and improved technical integrity. RTO RTO TIG. Following the SPE Forum Series in North America in July 2002, "Reconciling Real-Time Production Optimization with Reservoir Management," a volunteer group decided to foster and maintain a knowledge community to exchange ideas about the advancement of RTO. In June 2003, the group finalized the process for becoming an SPE TIG. The mission of the RTO TIG is to promote and encourage development of hardware and software tools including associated standards and work processes for RTO of hydrocarbon-production systems. Discussions and documents can be found at the SPE website. Registration is free, but SPE membership is required. Any member can post a discussing document. Discussions and documents are reviewed by a permanent program committee team to ensure quality and avoid commercialism. The first

deliverable of the TIG is the full-length paper, which establishes the key definitions pertaining to RTO, discusses available technologies, and establishes a preliminary definition of various classification levels for RTO. RTO Classification The initial focus of the TIG is real-time production optimization. Within this real-time production optimization, the classifications are viewed from the standpoints of "optimization cycle" and "automation network and time-scale hierarchy." This effort is similar to the multilateral-well classification that the completion engineers have created. Continuous Optimization Cycle. Continuous optimization cycle refers to a single generic process of optimization. Fig. 3 shows a continuous feedback loop describing the idea for reconciling real-time production optimization with reservoir management. This cycle fits in with a global field/asset plan for measuring operational changes (measure), continuous well- and reservoir-model update (interpret), defining optimal production strategy (optimize), and operating-parameters change for surface equipment and wells (control). Automation Network and Time-Scale Hierarchy. Monitoring and controlling the hydrocarbon-producing system, as in any other process plant, requires different timescale hierarchy. Basic Process-Control Level. The lowerlevel automation in the pyramid is the basic process-control level that should be appropriate to the dynamics of the process and can be a high-speed (measure and control) closed loop. Time scale is milliseconds, seconds, and minutes. Typical applications of basic process control can include the following. · A monitoring and control system of a crude, gas, and water separation process. · An automated wellhead unit that controls the gas lift injection rate. · A variable-speed pump controller.

Supervisory Control Level. The next level is the supervisory layer, usually closed loop. A typical application is a supervisory-control and data-acquisition system that interrogates remote terminal units for the supervision of remote hardware and instrumentation. Optimization Level. At the optimization level, certain techniques and applications are used to search for an optimum solution to determine the best exploitation scheme that maximizes value subject to some restrictions. A typical application is an automated well and field application that optimizes the well's gas lift injection value on the basis of a number of constraints and creates the set point for the lower basic process-control layer. Currently, it is usually a one-time process or an offline process. Integration Level. The integration level searches for optimum business-level solutions and integrates data from different problems. Because the time scale is on the order of months or years, the term planning or scheduling is more suitable than control. Levels of Classification. The purpose of classification is to help evaluation of existing or proposed systems and also to determine the availability of progressively advanced systems. It is similar in concept to the classification levels established by the group for technological advancements in multilateral wells. The objective is to construct the classifications on the basis of existing case histories. Tables 1 through 4 in the full-length paper list the levels of technology available in measurement, interpretation, optimization, and control. Conclusions Today, the oil industry has the technology and processes to achieve RTO of hydrocarbon-producing systems. There is a substantial economic incentive for RTO, but most operators may not be making effective use of it. RTO can increase production rates by as much as 3 to 10% and yield improvement in facility downtime as well as OPEX and CAPEX reductions. There are few examples of systematically applied RTO in the oil and gas industry. A new SPE TIG has been established to promote use of RTO. The TIG will focus on sharing technology, setting standards, and demonstrating real field examples of sucJPT cesses and failures.

For a limited time, the full-length paper is available free to SPE members at www.spe.org/jpt. The paper has not been peer reviewed.

DECEMBER 2003

35

P r o d u c t i o n / Fa c i l i t i e s

"Self-Lifting" Method to Eliminate Severe Slugging in Offshore Production Systems

Oil production from fields in water depths greater than 1800 m is a reality. The use of long deepwater risers that conduct production from multiple wellheads on the seafloor to the surface predisposes the system to severe slugging in the riser for a wide range of flow rates and seabed topography. Transferring the pipeline gas to the riser at a point above the riser base can reduce both the hydrostatic head in the riser and the pressure in the pipeline, consequently lessening or eliminating the severe slugging by maintaining the steadystate two-phase flow in the riser. Introduction Severe slugging can occur in two-phase-flow systems in which a segment or riser with upward slope follows a pipeline segment with downward slope. For such a system, at relatively low gas and liquid flow rates, liquid can accumulate at the riser base, blocking the gas flow. This situation will result in an increasing liquid level in the riser until the liquid reaches the riser top. Simultaneously, gas in the downward-sloping section will be compressed. When the gas pressure in the pipeline has increased enough to counteract the hydrostatic head of the liquid column, the gas will expand and push the liquid column violently out of the riser into the separator. Severe slugging will cause periods of zero liquid and gas production in the separator followed by very high liquid and gas flow rates. The resulting large pressure and flow-rate fluctuations and sudden surges in liquid production could cause the separator to overflow and shut down. Fluctuations in gas production could result in operational and safety problems during flaring, and the large pressure fluctuations could reduce the field's production performance and, ultimately, lead to reduced recoverable reserves. Three basic elimination methods have been proposed: backpressure increase, gas lift, and choking. The backpressure-increase method eliminates severe slugging by increasing the system pressure, which reduces the production capacity. In gas lifting, external gas is injected either into the riser or pipeline at the riser bottom to reduce the hydrostatic head in the riser or increase the gas flow rate in the pipeline. Gas lift equipment requires a large footprint on the platform and a large amount of gas. The operational cost of gas lift can be very significant. Choking increases the backpressure in proportion to the velocity increase in the riser. If the gas movement in the riser is stabilized before reaching the choke, steady flow will occur after a short flow period. Stabilization requires very careful choking to ensure minimum backpressure. Considering the dimensions of deepwater pipeline and riser systems, the severe-slugging phenomenon should be more pronounced, with possible occurrence at considerably higher system operating pressures compared with production systems at shallower depths. Therefore, system design and methods used to control or eliminate severe slugging become crucial when considering operation safety and the limited space on the platform. The objectives of the steady-state modeling study to prevent severe slugging in deepwater pipeline and riser systems were to predict the conditions under which the severe slugging could be eliminated by bypassing gas to the base of the riser as shown in Fig. 1, and to develop design criteria and procedures for field application. Physical Model In this system, gas and liquid enter the base of the riser through a downward-sloping pipeline. It is assumed that the length of this downward-sloping-pipeline section is sufficient to ensure fully stratified two-phase flow close to the riser base. Under normal (severeslugging) operation, gas and oil would proceed downward to the base of the riser, accumulate, and, after sufficient pressure is built up

This article, written by Technology Editor Dennis Denney, contains highlights of paper SPE 84227, "A Design Approach for `Self-Lifting' Method to Eliminate Severe Slugging in Offshore Production Systems," by J.Ø. Tengesdal, SPE, Pennsylvania State U. (now with Statoil ASA) and Leslie Thompson, SPE, and Cem Sarica, SPE, U. of Tulsa, prepared for the 2003 SPE Annual Technical Conference and Exhibition, Denver, 5­8 October.

in the gas "bubble," the mixture would be expelled to the surface through the riser. However, in the proposed self-lifting system, high-pressure gas is diverted from the pipeline and reinjected into the riser at a distance above the base. Above the reinjection point, the two phases are lifted to the surface separator. Criterion for Continuous Flow-- Simplified Model As a first approach to modeling the flow system, it is assumed that frictional pressure drops are negligible. Further, it is assumed that in the section of pipeline/riser/separator system under consideration, pressure and temperature variations are small enough that gas and liquid pressure/volume/temperature and flow properties (e.g., densities and holdups) can be considered as approximately constant. Finally, the gravitational pressure drop in the gas bypass is ignored. In the simplified model shown in Fig. 2, there is no pressure drop between Points A and B, regardless of whether the liquid or gas legs are considered. Under this assumption, the maximum height above the riser base of the gas injection point is obtained when the nose of the bubble is at the gas takeoff point. In practice, there will be some pressure drop in the gas leg, so even if the gas takeoff point (A) and injection point (B) are at the same elevation, the nose of the gas bubble will lie between point A and the riser base. Having fixed the positions of the takeoff and injection points, a simple pressure balance shows that continuous flow would occur if the injection pressure at Point B is sufficient to overcome the pressure losses in the remainder of the riser.

Fig. 1--Schematic of self-lifting slug-elimination technique.

36

DECEMBER 2003

Fig. 2--Simplified model of pipeline/riser system.

Criterion for Continuous Flow-- Rigorous Model In practice, there will be a pressure drop in the gas-bypass leg (i.e., between Points A and B in Fig. 2) that will influence the optimal location of the injection point. The gas-bypass system will operate successfully as long as the nose of the penetrating gas bubble lies between the gas takeoff point and the riser base. The minimum allowable pressure drop in the gas leg will occur when the nose of the gas bubble is at Point A, and the hydrostatic pressure drop in the liquid leg is determined by the liquid height difference. If the pressure drop in the gas leg were to fall below this level, the nose of the gas bubble would recede to a point above the gas take-off point (A), and liquid would enter the gas bypass. At the opposite extreme, the maximum allowable pressure drop in the gas bypass occurs when the nose of the penetrating gas bubble is at the riser base. If the pressure drop in the gas line increases further, the gas bubble would circulate around the bottom of the riser and aerate the liquid. Under this mode of operation, gas would enter the riser through the path of minimum resistance. A successfully operating system would exhibit a bypass pressure drop between the minimum and maximum allowable pressure drops. Model Performance The steady-state model provides the range of minimum and maximum pressure losses that can be tolerated through the gas bypass to ensure stable steady-state operation of the system. Numerous sets of steady-state data were collected for pipeline sections with (downward) slopes of -1, -3, and -5°. Comparisons were made of the experimental steady-state pressure drops with the pressuredrop ranges predicted by the model for downward pipeline inclination angles of -5, -3, and -1°, respectively. For the -5° case,

the actual measured maximum gas-bypass pressure drop was 1.21 psi and the average was 0.56 psi. All the measured data (161 points) for this case fell within the bounds predicted by the steady-state model. The maximum and average gas-bypass pressure drops for the -3° case (155 points) were 1.36 psi and 0.83 psi, respectively. In this case, five measured points had pressure drops greater than those predicted by the model, and one point had a less-than-predicted pressure drop. Finally, for the 114 measured data points in the -1° case (with maximum and average gas-bypass pressure drops of 0.82 psi and 0.42 psi, respectively), 28 measured values fell below the predicted range with none above. Overall, the steady-state model performed very well in bracketing the pressure losses observed in the gas bypass. It is important to note that the pressure losses in the self-lift system are very small. Successful system operation will impose a minimal backpressure on the production facilities upstream of the riser and, therefore, will have a minimal negative effect on reservoir productivity. Also, it is difficult to measure pressure drops across the bypass accurately. It is believed that a major cause of the outlier data was the difficulty in obtaining precise measurements. In partic-

ular, the worst case was for a pipe-slope angle of -1°, for which the operational pressure drops were very small. Conclusions A steady-state model was developed that predicts the acceptable range of pressure drops across the gas-bypass line for successful operation of the self-gas-lift severe-slugging-elimination method. Both model and experimental data demonstrate that the pressure drops incurred by use of the gas bypass are very small (less than 2 psi for the experimental facility used to test the concept), which suggests that the effect of implementing the system on overall productivity will be minimal. Comparison of the operational ranges predicted by the model with actual measured steady-state gas-bypass pressure drops was acceptable. It was hypothesized that practical difficulties in measuring the small gas-bypass pressure drops were responsible for any discrepancies between JPT measured and model data.

For a limited time, the full-length paper is available free to SPE members at www.spe.org/jpt. The paper has not been peer reviewed.

STATEMENT OF OWNERSHIP MANAGEMENT AND CIRCULATION (Required by 39 U.S.C. 3685). 1. Title , of publication, Journal of Petroleum Technology. 2. Publication No. 0028-1960. 3. Date of filing, 24 September 2003. 4. Frequency of issue, monthly. 5. No. of issues published annually, 12. 6. Annual subscription price, $45. 7. Complete mailing address of known office of publication, SPE Inc., P Box 833836, 2222 Palisades Creek .O. Dr., Richardson, TX 75080, Dallas County. 8. Complete mailing address of the headquarters or general business offices of the publishers, SPE Inc., P Box 833836, Richardson, TX 75083-3836. 9. Name and address of pub.O. lisher, Georgeann Bilich, P Box 833836, Richardson, TX 75083-3836. Name and address of editor, John .O. Donnelly, P Box 833836, Richardson, TX 75083-3836. 10. Owner, Society of Petroleum Engineers (SPE) Inc., .O. P Box 833836, Richardson, TX 75083-3836. 11. Known bondholders, mortgagees, and other security holders .O. owning or holding 1 percent or more of total amount of bonds, mortgages, or other securities (none). 12. The purpose, function, and nonprofit status of this organization and the exempt status for Federal income tax purposes have not changed during preceding 12 months. 13. Publication name: Journal of Petroleum Technology. 14. Issue date for circulation data below: September 2003. 15. Extent and nature of circulation:

Average Number Copies Each Issue During Preceding12 months A. B. Total number copies (net press run) Paid and/or requested circulation 1. Paid/requested outside-county mail subscriptions stated on Form 3541 2. Paid in-county subscriptions 3. Sales through dealers and carriers, street vendors, counter sales, and other non-USPS paid distribution 4. Other classes mailed through the USPS Total paid and/or requested circulation Free distribution by mail (samples, complimentary, and other free) Free distribution outside the mail (carriers or other means) Total free distribution Total distribution Copies not distributed 53,201 Actual Number Copies of Single Issue Published Nearest Filing Date 55,118

27,748 none 23,186 none 50,934 0 548 548 51,482 1,719 53,201 98.9%

28,270 none 25,598 none 53,818 0 840 840 54,658 460 55,118 98.5%

C. D. E. F . G. H.

I. Total Percent paid and/or requested circulation

17. I certify that the statements made by me above are correct and complete. Georgeann Bilich, publisher.

DECEMBER 2003

37

Information

Production/Facilities Overview

10 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate

652703