Optimizing Horizontal Well Performance in Nonuniform Pressure Environments Using Passive Inflow Control Devices

Authors: Dr. Dimitris Krinis, Drew Hembling, Naseem J. Al-Dawood, Sadiq A. Al-Qatari, Dr. Sam Simonian and Giovanni Salerno

ABSTRACT
Passive Inflow Control Devices (ICDs) are used to enhance performance of horizontal wells in unfavorable environments, such as nonuniform permeability and/or pressure along horizontal sections. This is the first attempt at using ICDs to manage inflow along horizontal wells with a substantial reservoir pressure differential. This article presents how passive ICD completion technology is used to optimize well productivity and reduce water production in a well with significant reservoir pressure differential across the horizontal section. Prior to running the ICDs, this horizontal carbonate well exhibited a nonuniform production profile, cross-flow and high water production, resulting in reduced well performance. A production log of the initial (open hole) completion confirmed the downward cross-flow of fluids from heel to toe and production contribution from only the first 10% of the horizontal section. Following the installation of the ICD system, water production was greatly reduced and the oil rate doubled, indicating significant improvement of the well performance. A second pipeline terminal (PLT) survey showed contribution from the entire horizontal section and the elimination of the cross-flow. The key factor in the success of this project was the ICD system design derived from numerous wellbore hydraulic simulations, which determined the appropriate number of ICD units and compartments.

reservoir pressure was 200 psi from heel to toe, with highpressure at the heel.

BACKGROUND
Horizontal wells are common practice in this giant Middle Eastern field with a production history of more than 50 years. The reservoir is a composite Jurassic carbonate anticline; with several gently dipping crestal regions containing undersaturated Arabian Light grade oil. Full pressure maintenance was initiated in the 1980s by peripheral power water injection. Infill drilling is still underway, tapping oil behind the flood front areas utilizing horizontal well technology. The most prolific zones in this gravity dominated reservoir exhibit high matrix (300 millidarcies (md) - 2,000 md) permeability, which is the principal source of production. The subject well was drilled and completed open hole in 2005. The total depth was 13,125 ft with a reservoir section of 5,200 ft. The well trajectory is illustrated in Fig. 1. The horizontal lateral was placed at the top of the reservoir in an area with an oil column ranging from 60 ft - 100 ft underlain by swept intervals. The direction in which this well was drilled favored considerable reservoir pressure gradients as a result of the peripheral water injection. The logging while drilling (LWD) interpretation showed a high quality homogeneous oil bearing formation. The initial rate of this well was 11,000 barrels per day (BPD) with a 4.6% water cut. The well experienced rapid oil rate decline and subsequent increase in water cut. The first open hole PLT survey, conducted in late 2007, showed only the first 500 ft of the horizontal section contributing. There was no contribution from the remaining 4,700 ft of the lateral. The first PLT survey included both a flowing and a shut-in pass. During the shut-in pass, considerable downward cross-

INTRODUCTION
Horizontal wells have been used globally to maximize reservoir contact and ultimately improve well performance. Variations in permeability, pressure and frictional pressure drop along horizontal sections can create nonuniform inflow profiles, which can lead to early breakthrough of water or gas in reservoirs with an aquifer and/or gas cap. Breakthrough typically occurs in regions of high permeability or at the heel of the lateral section in homogeneous reservoirs. This ultimately leads to a reduction in reserve recovery and well performance. The use of Inflow Control Device (ICD) systems to effectively manage the nonuniform inflow profile has been described in past publications1-4. In this article, ICDs are used to control nonuniform pressure profiles. The variation in
44
SPRING 2009 SAUDI ARAMCO JOURNAL OF TECHNOLOGY

MD (ft)
6,750 6,800

TVD (ft)

6,850 6,900 6,950 7,000 7,500 8,500 9,500 10,500 11,500 12,500 13,500

Fig. 1. Well trajectory Reservoir section.

flow was evident, Fig. 2, which is the direct effect of the differential reservoir pressure along the horizontal section, about 200 psi from heel to toe (heel being at a higher pressure). Due to coil tubing limitation, to reach the entire lateral, only one thief zone could be identified. Figure 2 also shows the production profile of the flowing pass. Again the downward cross-flow is observed with the water flowing to the bottom of the lateral, whereas oil is flowing upward and being produced to surface. This counter-current flow is clearly visible in the wellbore cross-section velocity profile as derived by the multiple spinners of the logging tool (Fig. 2, Insert) at location A. The oil rate measured during the flowing conditions was 4,000 reservoir barrels per day (RBPD) with a 30% water cut. Due to the significant reservoir pressure gradient and the fact that well performance was declining, it was decided to deploy a passive ICD completion to regain well performance by increasing contribution across the entire lateral and decreasing water cut.

Fig. 2. Formation porosity, Shut-In and Flow Profile of open hole completion. Insert: Cross section Velocity Profile at location A of the wellbore.

Fig. 3. ICD completion with swell packers.

ICD COMPLETION STRING DESIGN

A passive ICD completion was installed into the existing horizontal lateral in early 2008 without deepening or sidetracking the well. The objectives of the ICD completion were the following: (a) enhance production contribution from the entire 5,200 ft lateral, (b) overcome the unfavorable heel to toe differential pressure, thus reducing the downward cross-flow, and (c) minimize the water influx to prolong the well life. Numerous wellbore hydraulic simulation runs were carried out to determine the number of ICD units and segments required. Sensitivity runs to various fluid rates ensured that the ICD completion would satisfactorily function within the preferred flow range and maintain positive contribution across the entire lateral. A commercial wellbore hydraulic simulator, NETool, was used for this exercise. An optimum design rate was selected at 8,000 barrels of fluid (BFD) (oil and water). The design of the ICD completion is given in Fig. 3. Swellable packers were used to create the different compartments/segments across the lateral. The number of nozzle types for the ICD units and their placement were determined through the NETool simulation runs with the above objectives.

Fig. 4. Well performance test flow data.

Fig. 5. ICD vs. open hole completion inflow profile.

ICD COMPLETION PERFORMANCE

The well was worked over and re-completed with 22 ICDs and seven swellable packers. Post re-completion, surface well test data indicated stable oil production with reduced water cut (< 10%). Figure 4 illustrates the pre- and post-ICD completion flow data. These flow rates suggest that the ICD completion was performing as expected. Note that the ICD completion was intensively tested at both target and elevated rates to ascertain its performance and coning tendency.

To confirm the inflow profile along the length of the ICD completion, a PLT survey was conducted in mid 2008. Shut-in and three flowing passes (high, medium and low rate) were run to both assess the performance of the ICD completion at different rates, and also determine the optimum rate. Figure 5 shows the PLT inflow profile at the medium test rate. The cross-flow, evident in the open hole log, was eliminated by the ICD completion while achieving contribution from the entire lateral. Furthermore, the actual profile compared favorably with the predicted profile as per design of the completion (Fig. 5, red line) at the same rate.

ICD MODELING
Additional simulation runs were made using the wellbore hydraulic simulator to match both flow and pressure profiles. Figure 6a illustrates both the simulated and the measured (by the PLT survey) inflow profiles along the reservoir interval for the medium rate case. The characteristic stair-step profile is
SAUDI ARAMCO JOURNAL OF TECHNOLOGY SPRING 2009

45

seen in both sets of data at the depths where the ICD modules are located. Overall, the simulated profile compares well with the actual profile. The middle section of the horizontal lateral was found to be the most challenging to match. Figure 6b shows the match of the flowing bottom-hole pressure (FBHP) obtained from the PLT survey for the ICD completion. To achive this match the expected (green curve) reservoir pressure profile had to be adjusted to a profile with reduced differential pressure between the heel and the toe (magenta curve). In addition, a variable skin along the lateral was introduced to further enhance the FBHP and rate match. It should be noted that there is no unique solution to this set of data. Figure 6b also shows the simulated annular pressure profile. Furthermore, a simple mathematical model was developed, based on Bernoullis Eqn. 1, to calculate the pressure drop across a nozzle-type ICD. A series of flow loop tests were conducted by the ICD vendor using commercially available oil to derive the actual performance (pressure drop vs. rate) curve of the specific ICD module deployed. The experimental data

was also employed to calibrate the mathematical model and derive the value of the flow coefficient (Cd) as shown in Eqn. 5. This equation is an extension from Eqn. 3, based on Venturi nozzles, and the flow coefficient accounts for the deviation from the ideal nozzle. Figure 7 shows a comparison plot between the flow loop test data (blue curve) and the mathematical model for single phase oil (red curve). The mathematical model is in close agreement with the experimental data indicating that it can reliably predict the nozzle-type ICD performance. Thus, this model was used to predict the performance of the ICD at the actual reservoir properties (downhole oil density and viscosity) applicable to this case study well. The calculated pressure drop vs. rate (green) curve is also plotted in Fig. 7, which exhibits reduced pressure drops due to the lower density of this particular oil. The total pressure drop across a system is given by:

(Bernoullis Eqn.)

(1)

PLT vs. Simulated Flow Profiles

14
ICD #: 1 1 2 2 3 4 4 5

PLT vs. Simulated Pressure Profiles

3150

(Medium Choke) Reservoir Pressure Matched Profile

Flow Rate (1000 RB/D)

12 10 8 6 4 2 Depth (ft) 0 8000 9000

Actual (PLT) Rate Simulated (NETool) Rate As Per Design

Pressure (psi)

(Medium Choke)

3050

2950

Simulated Annular Pressure

2850

Reservoir Pressure Starting profile Simulated FBHP

Actual FBHP

2750

2650

Depth (ft) 10000 11000 12000 13000

2550 8000 9000 10000 11000 12000 13000

Fig. 6a and 6b. PLT and NETool simulated data vs. depth comparison at medium choke.

Experimental vs. Mathematical Model Data

300

Pressure Drop Across ICD (psi)

Experimental Data at Test Conditions 250 Model at Test Conditions 200 Model at Case Study Reservoir Conditions

150

100

50 Rate (B/D) 0

100

200

300

400

500

600

Fig. 7. Experimental test data vs. simple mathematical model.

46

SPRING 2009 SAUDI ARAMCO JOURNAL OF TECHNOLOGY

Where (Potential Energy Term) (2)

REFERENCES
1. Krinis, D., Al-Dawood, N. and Soremi, Y.: Horizontal Well Performance Optimization using Linearized Inflow Control Devices, SPE paper 117213, presented at the 2008 Abu Dhabi International Petroleum Exhibition and Conference, Abu Dhabi, U.A.E., November 3-6, 2008. 2. Raffn, A.G., Zeybek, M., Moen, T., Lauritzen, J.E., Sunbul, A.H., Hembling, D.E. and Majdpour, A.: Case Histories of Improved Horizontal Well Cleanup and Sweep Efficiency with Nozzle-Based Inflow Control Devices in Sandstone and Carbonate Reservoirs, OTC. 3. Hembling, D.E., Sunbul, A.H. and Salerno, G.: Advanced Well Completions Result in Enhanced Well Productivity and Recovery in Saudi Aramcos Offshore Fields, 2007 SPE Asia Pacific Oil & Gas Conference and Exhibition, Jakarta, Indonesia, October 30 - November 1, 2007. 4. Al Qudaihy, D.S., Al Qahtani, H.B., Sunbul, A.H., Hembling, D.E. and Salerno, G.: The Evolution of Advanced Well Completions Results in Enhanced Well Productivity and Recovery in Saudi Aramcos Offshore Fields, IADC/SPE Asia Pacific Drilling Technology Conference and Exhibition held in Bankok, Thailand, November 13-15, 2006.

(Kinetic Energy Term)

(3)

(Frictional Loss Term)

(4) (5)

Where h: Height/Elevation g: Gravity Acceleration Constant : Liquid Density V: Fluid Velocity Q: Liquid Rate : Constant L: Length d: Nozzle Diameter : Friction Factor Cd: Flow Coefficient For nozzle type ICD modules, the most dominant term is the kinetic energy, Eqn. 3. Furthermore, this model was used for quality assurance, using the results from the hydraulic simulator, and expanded the learning curve and understanding of the pressure drops associated with evaluating ICD completions.

CONCLUSIONS
This article presented a case history showing how passive ICD completion technology can be used to effectively manage inflow profiles, even under adverse reservoir conditions. This horizontal well had a nonuniform reservoir pressure across the horizontal well, leading to considerable cross-flow, uneven production and increased water cut; resulting in reduced well performance. It has been demonstrated that a well-designed ICD completion, assisted by wellbore hydraulic modeling can accomplish the following: Promote production from the entire horizontal lateral, and mitigate the effect of the severe pressure gradients, eliminating cross-flows. In this well, the ICD completion reduced water cut from 30% to less than 10%, significantly improving well performance.

ACKNOWLEDGEMENTS
The authors wish to thank the management of Saudi Aramco and FloTech for actively supporting this work and granting permission to publish this article.

SAUDI ARAMCO JOURNAL OF TECHNOLOGY

SPRING 2009

47

BIOGRAPHIES
Dr. Dimitris Krinis is a Petroleum Engineering Consultant in the Abqaiq Reservoir Management Division of the Southern Area Reservoir Management Department. He graduated with a B.S. and M.S. degree in Mining Engineering from the National Technical University of Athens, Athens, Greece in 1985 and then received his Ph.D. degree in Petroleum Engineering from Heriot-Watt University, Edinburgh, Scotland in 1989. Before joining Saudi Aramco in 2000, he worked as Reservoir Engineer for AFPC in Syria. For the first 12 years of his career, Dimitris worked for Marathon Oil and Shell in the North Sea and Malaysia as a Petroleum Engineer involved in various projects, including field development, reservoir simulation, fluid characterization, well testing and marginal oil and gas prospect evaluation. Drew Hembling is a Completion Team Leader working in the Petroleum Engineering Support Division of Saudi Aramcos Producing Facilities Development Department. He graduated from West Virginia University, Mogantown, WV in 1983 with a B.S. in Petroleum Engineering. Drew worked for Conoco Inc. from 1983 to 1997 and ARCO in Anchorage, AK from 1997 to 2000. He joined Saudi Aramco in 2000 as a Completion Specialist. Naseem J. Al-Dawood is a Supervisor in the Abqaiq Reservoir Management Division of the Southern Area Reservoir Management Department. In 1990 he received his B.S. in Petroleum Engineering and in 1992 he received his M.S. degree in Petroleum Engineering, both from the University of Alabama, Tuscaloosa, AL. He joined Saudi Aramco in 1993 and has worked in various disciplines including reservoir management, reservoir description, drilling and production engineering.

Sadiq A. Al-Qatari received his B.S. degree in Petroleum Engineering from the University of Southern California, Los Angeles, CA. He joined Saudi Aramco in May 1981 and since then Sadiq has been involved in all kinds of operations as a production, drilling, log interpreter and reservoir engineer. Currently, he works as a Petroleum Specialist in the ANDR field. Dr. Sam Simonian is the Technical Director for FloTech Ltd. He received a M.E. in Aeronautical Engineering from The City University, London, United Kingdom in 1988 and received a Ph.D. in Multiphase Fluid Dynamics in 1993. Before joining FloTech Ltd. in 2006, he worked for Schlumberger Wireline, Anadrill and finally, Well Completions and Productivity for 15 years. Thereafter, Sam joined Eclipse Petroleum Technology as a Reservoir Engineer/Production Technologist for 3 years. He worked with various oil and gas operators (e.g., Shell, BHP, BG and BP) worldwide, specializing in production optimization, field development planning, reservoir simulation and well testing. Giovanni Salerno is the General Manager of FloTech Ltd. He received his B.S degree in 1995 in Petroleum Engineering from the University of Bologna, Bologna, Italy. Before joining FloTech Ltd., in 2008, he worked for 6 years with an International Operator, ENI, as a Drilling and Completion Engineer, gaining experience in stimulation, production optimization, well construction, sand control and completion operations for both onshore and offshore fields. Thereafter, Giovanni spent 7 years with Baker Oil Tools as a Completion Specialist, mainly focused on completion solutions for the Middle East, based in Saudi Arabia. He has also worked in Europe, Africa and the Far East.

48

SPRING 2009 SAUDI ARAMCO JOURNAL OF TECHNOLOGY