Improving performance in unconventional oil reservoirs

The recovery efficiency of unconventional oil reserves is very low, due to micro-Darcy formation permeability and rapid depletion of pore pressure near to the fractures and wellbore. Several enhanced oil recovery (EOR) techniques have been suggested, to increase production efficiency in these reservoirs. Two examples promote injection of gas or carbon dioxide (CO₂) in fractured wells, and either producing oil from adjacent wells or back-producing the injected gas and reservoir fluids in the same wellbore after a suitable soaking period (huff & puff).

The effective distribution of the injected gas, and the ability to keep gas in the reservoir for maintaining energy and/or having longer contact with interstitial oil, can significantly improve the performance of the EOR process in unconventional oil wells. This can be achieved via the use of advanced completions utilizing passive inflow control devices (ICDs) and autonomous inflow control devices (AICDs).

HUFF & PUFF PROCESS

To produce a significant amount of the remaining oil in place, the development of effective EOR techniques is necessary. Due to low injectivity and poor sweep efficiency, conventional secondary recovery scenarios, such as waterflooding, are ineffective in unconventional reservoirs. While alternate injection concepts have been suggested,^1,2 gas huff & puff appears to be the most practical technique. It comprises three steps: injection, soaking and production. For instance, EOG Resources³ disclosed oil recovery improvement of 30% to 70% from Eagle Ford shale wells by injecting natural gas, using huff & puff techniques.

Simulation studies have focused on providing a better estimation of the added value of gas injection methodology and gas injection optimization. Research in 2014 and 2016 evaluated the impact of the reservoir uncertainties, number of cycles, injection time, injection rates and primary depletion period.^4,5 It concluded that the fracture length and conductivity are less important than the other parameters for successful huff & puff injection.

Broadly speaking, economics favor shortening the injection and soaking times, and increasing the production time. In addition, it is desirable to keep the injected gas in contact with the reservoir matrix, even during the production phase, by restricting the production of free gas that has not had a chance to interact with the interstitial oil.

ADVANCED COMPLETIONS

Extended-reach wellbore architecture (high-angle, horizontal and multi-lateral wells) delivers more efficient exploitation of complex geologic structures and provides enhanced reservoir contact for layered, laminated and compartmentalized formations. However, it also has brought new challenges, such as uneven inflow from the reservoir toward the wellbore. Such challenges are the result of a variation in the reservoir properties. This can lead to early water and gas breakthrough in one region of the horizontal wellbore, adversely affecting oil production in the rest of the wellbore: potentially forcing a well shut-in.

By managing the inflow and outflow of fluids from/to the reservoir, advanced completions can respond to changing conditions throughout the life of the well. These technologies come in three forms:

• Interval Control Valves (ICV) – active flow control devices
• Inflow Control Devices (ICD or FCD) – passive flow control devices
• Autonomous Inflow Control Devices (AICD or AFCD).

Each requires segmentation of the wellbore and isolation of the sections by annular flow isolation tools, including swell packers, mechanical packers or seal bore stacks. This article focuses on passive and autonomous flow control devices.

Passive flow control device (ICD). Passive ICDs incorporate a variety of elements to generate a pressure drop as a function of the flowrate going through the device. For instance, the simplest, most common types of ICD use a nozzle or orifice.

Fig. 1. ICD unit mounted into sand screen joints, illustrating the injection flow path. ©Tendeka

Passive ICDs are used in long horizontal water injection wells to balance the injection rate into the reservoir along the length of the well. One or more ICD devices are mounted on the base pipe of a screen joint, and other than at the housing of ICDs, the base pipe is unperforated. Multiple ICD screen joints are normally run in a horizontal well with sections isolated by swell packers or mechanical packers between the main conduit and the wellbore.

As shown in Fig. 1, the flow travels down the injection conduit (screen base pipe), and through the ICDs in each joint, to the housing at the top of the screen. Here, the flow is redirected from a radial direction to an axial direction. The flow continues down the annular space between the screen and the base pipe, and it goes out the screen to the annular space between the screens and the wellbore, and then into the reservoir.

Figure 2 shows a section view of an ICD device for injection that incorporates the ability to prevent flow in the production direction (from the annulus to the production conduit). This type of ICD is required to employ flow control in a huff & puff EOR process.

Fig. 2. Bypass injection ICD device with check capability in the production direction. ©Tendeka

Autonomous inflow control device (AICD). An AICD is a flow control device, in which the pressure drop of fluid flowing through the device is dependent on the composition of the fluid or other properties, such as viscosity, density, or flowrate. The fluid-dependent pressure drop can be created by changing the geometry of the flow restriction or the direction of the flow path of the fluid as a function of the controlling property.⁶

The AICD provides additional restriction to unwanted fluids, such as water or gas, and creates the additional restriction without any connection to, or remote actuation from, the surface and without any intervention by the operator. When installed in a well that has been segmented into multiple compartments, this device prevents excessive production of gas after breakthrough occurs in one or more compartments.

A popular AICD design is the Rate Controlled Production (RCP) valve. Tendeka has deployed more than 42,000 of these devices in over 280 wells worldwide. Figure 3 illustrates the principle components of the RCP AICD.

The RCP AICD creates a variable, rate-and-fluid-dependent pressure drop, based on the size of the gap between a free-moving disk and the top plate of the housing in which it is contained. Flow enters the device through the nozzle in the top plate of the body. This impacts the disk and disperses radially through the gap between the disk and the top plate. It then moves around the top plate and is discharged through several outlet ports in the body, Fig. 4.

Fig. 3. Construction of an RCP-type AICD. ©Tendeka

The RCP AICD is typically incorporated as part of a screen joint, as shown in Fig. 5. The produced fluids enter the completion through the screen and flow, in the annular space between the screen and the unperforated base pipe, into the AICD housing, where the device is mounted. Fluids then flow through the device into the interior of the production conduit where they combine with the flow from other zones and devices. The RCP AICD functions as a check valve in the injection direction.

The design and performance of the RCP AICD make it well-suited for installations in long horizontal wells and multi-laterals with high degrees of reservoir heterogeneity, and uncertainty regarding future well conditions. The ability to change its performance as a function of the fluid flowing through it makes it ideal to control gas influx in reservoirs with various oil columns or reservoirs that have high gas-oil ratios and saturated oil.^7,8,9

REDUCING UNCERTAINTY AND IMPROVING PERFORMANCE

Fig. 4. Flow path inside an RCP AICD valve ©Tendeka

To illustrate the application of advanced completions in a huff & puff gas EOR well, in a shale oil reservoir, consider a horizontal well drilled in an unconventional oil formation and completed as either open-hole or with perforated, cemented casing. Several hydraulically induced fractures are then created in the reservoir.

The new completion is composed of an internal liner with annular flow isolation packers to generate multiple hydraulically isolated segments in the wellbore. Each isolation packer is placed to compartmentalize either individual fractures or clusters of hydraulic fractures, previously induced, outside the internal liner. A production packer or liner hanger/packer can be installed at the top of the internal liner to isolate the formation from the upper wellbore, Fig. 6. 

Each segment comprises one or more flow-checking ICDs and one or more flow-checking AICDs. The number and size are determined by the expected prevailing operating conditions to control both gas injection and oil and gas production at specified rates.

The well is initially produced for some time to deplete the pressure in the reservoir, proximal to the wellbore and fractures, and for the economic benefit of early oil production. Once the production rate and pressure have declined, gas injection into the wellbore is initiated.

Fig. 5. Flow path of production fluid through an AICD mounted on a sand screen joint. ©Tendeka

The well is then closed for an appropriate period to allow the gas to diffuse into the interstitial oil in the formation’s pore-spaces, as well as the solvent-oil solution dispersing back into the micro-fracture labyrinth. After this period, the well is opened to production again, with gas and oil produced from the formation into the micro-fractures and hydraulically induced fractures, then on to the wellbore, passing through the AICDs into the production conduit.

As the properties of formation and fractures for each segment are different, the segments producing large gas volume fractions are restricted by the AICDs. Simultaneously, the AICDs at segments producing high oil volume fractions will impose a little restriction on the oil flow. This control retains more gas in the reservoir to diffuse further and to maintain reservoir pressure and energy to enhance instantaneous and ultimate oil recovery. The cycles of injection and production are repeated multiple times, to increase the recovery of hydrocarbons from the well/reservoir.

In summary, under normal circumstances with conventional completions, injected gas or CO₂ flowing back would be preferentially produced, due to the favorable mobility of these fluids during the production stage of a huff & puff cycle. However, the AICD provides greater flow restriction to gas and CO₂ than to oil, and as such, fractures zones dominated by gas phase (or CO₂) are subjected to a very high pressure drop, while the zones dominated by oil are produced with a minimum restriction. This facilitates maintaining pressures in the zones that the gas (or CO₂) has not had enough time to react with the oil successfully, while maximizing oil production from high oil-bearing fracture zones.

Fig. 6. Proposed advanced completions, including flow control devices in a multiple-fractured horizontal well. ©Tendeka

As demonstrated by simulations, Tendeka believes the duration of the soaking period can be shortened with AICDs, and the effectiveness of the gas or CO₂ injected can be increased significantly.

A BAKKEN-LIKE TIGHT OIL RESERVOIR SIMULATION STUDY

The Bakken formation is located between the Devonian Three Forks and Mississippian Lodgepole Limestone formations. In addition to multi-stage, fractured horizontal wells, EOR is crucial to increasing the recovery factor and maximizing production from the fields. For instance, several studies have revealed the possibility of increasing the oil recovery factor by 10% to 15%, if EOR is applied.

To demonstrate the method in a pessimistic scenario, a model with highly heterogeneous, static geological properties was built. The static and dynamic properties used to evaluate the performance of huff & puff gas injection are either identical or within the same order of magnitude to the Bakken field.

A 3D compositional box model was built, using 300*200*10 active grid cells, with the size of 40*40*3ft in X, Y and Z directions, respectively, to capture the minor differences when it came to simulation. Porosity was then populated, using the average field porosity of 27% as a reference, by creating a custom porosity log in LAS format, using numbers generated by a random number generator.

The log was then upscaled to the grid, using arithmetic averaging and the data correlated in an isotropic variogram. The data was utilized to interpolate the porosity, using Sequential Gaussian Simulation. This provided a heterogeneous and random porosity distribution with the limited data available. Permeability in the X, Y and Z directions was then populated, using Poro-Perm correlations, allowing the X and Y permeabilities to a range between 0.016mD and 0.046mD, and permeability between 0.011mD and 0.032mD in the Z direction. The effects of geomechanics and dual porosity-dual permeability are not included in this study.

Well design and structure. A horizontal well with reservoir contact length of 5,600 ft was defined in the center of the model, to allow consistent boundary effects from all sides of the model. Six stages of fracture were placed to improve oil production. The heterogeneity in hydraulic fractures was induced by randomly spacing the fractures along the horizontal section of the wellbore. The fracture length was also varied.

As the focus was on the lower completion, the upper completion was not modelled. A 5.5-in. production liner was inserted into the well, to the toe of the well.

RESULTS

To include huff & puff EOR, the cyclic CO₂ injection was performed for three years. Two MMscfd of CO₂ were injected into the well, with a maximum bottomhole injection pressure of 5,750 psi. Injection started right after the first two years of natural depletion. The injection comprised 12 cycles of one-month gas injection and two months of back production of oil for each cycle (four per year). The well then continued to produce under natural depletion for a further five years.

It was observed that if no EOR was applied to the reservoir, a low primary recovery factor would be achieved. The results also show that gas injection is suitable for the tight environment, as total production has increased significantly, from 100,000 to 153,000 bbl until the end of injection time, or increased from 140,000 to 165,000 bbl as the ultimate total oil production over ten years.

Moreover, two different completions—conventional and advanced—were investigated to improve the recovery from this reservoir, when gas huff & puff is applied. The volume of injected gas (720 MMscf), and the ultimate well production time (ten years), were assumed to be fixed for all the cases, to properly compare different scenarios.

The advanced well completion’s design is based on the location of the hydraulic fractures. Therefore, the well is segmented into six zones by the placement of five packers. This allows control of individual fractures through the installation of 12 5-mm AICDs (two at each zone). The wellbore also has six bypass injection valves (one at each zone), to allow the injection of gas from inside the tubing to the reservoir.

A uniform design of the AICD completion is used to optimize the well performance. However, the oil rate production profile from the well with AICDs is superior, compared to the profile from the conventional completions, where the same total volume of CO₂ (720 MMscf) is injected into the well.

The study found total oil production from the well could be increased significantly, using advanced completion methods, while decreasing total gas production from the well.

For example, the cumulative produced oil increases 37% with EOR and an AICD completion, together, compared to 18% with EOR and a conventional completion. AICD completions reduced total gas production 6% over the lifetime of the well and, notably, during the three years of the huff & puff process. These results are due to AICDs keeping more gas in the reservoir. An increased reservoir pressure near the well is achieved, while the injection pressure is almost constant throughout the injection periods.

AICD completions also keep CO₂ in contact with oil for a longer period, resulting in more exchange of components between oil and CO₂ and, subsequently, decreased oil density. The autonomous devices, therefore, balance the production from each zone.

Notably, the study shows that the well with a conventional completion requires the injection of more volumes of gas (~2 times) than the AICD well, to pressurize the well and deliver the target pressure. Economically, this saves a huge injection cost for the advanced completion method.

Thus far, this process has been applied to a highly heterogeneous model, with varying fracture lengths along the wellbore. A small uncertainty study was also undertaken to analyze how the advanced completions would perform in models with varying initial reservoir pressure, fracture lengths and huff & puff schedules, compared to conventional completions. Here, the AICD completion design was kept the same during the uncertainty analysis, but it can be tailored to deliver optimum performance.

CONCLUSIONS

The study by Tendeka has demonstrated how advanced completion technology  can be applied to balance the distribution  of gas injection along the length of the wellbore and help control the early back-production of gas in a huff & puff gas EOR process for unconventional oil recovery.

ICDs and AICDs appear to improve the recovery efficiency of the huff & puff gas EOR process while mitigating the impact of reservoir uncertainties. Research also shows that advanced completions can deliver added value to a huff & puff project, if multiple miscibility contacts are encountered, because the advanced completion keeps the gas in the reservoir, allowing it to develop miscibility with interstitial oil.

While increased cumulative oil production can be achieved with advanced completions, the use of the completion minimizes the requirement of long soaking periods and reduces the volume of injection gas required. Hence, more economical production can be achieved from the huff & puff process. 

REFERENCES

1. MacPhail, W. F. P., and W. F. J. Deeg, “Well injection and production methods, apparatus and systems,” U.S. Patent Application 15/222090, 2017.
2. Dombrowski, R. J., et al., “Fluid Injection in Light Tight Oil Reservoirs,” U.S. Patent Application 13/781185, 2013.
3. Rassenfoss, S., “Shale EOR works, but will it make a difference,” Journal of Petroleum Technology, Vol. 69, No. 10: 34 – 40, 2017.
4. Yu, W., H. Lashgari, and K. Sepehrnoori, “Simulation study of CO2 Huff-n-Puff process in Bakken tight oil reservoirs”, SPE doi:10.2118/169575-MS, April 17, 2014.
5. Yu, Y., L. Li, and J. J. Sheng, “Further discuss the roles of soaking time and pressure depletion rate in gas huff-n-puff process in fractured liquid-rich shale reservoirs, “ SPE paper doi:10.2118/181471-MS, Sept. 26, 2016.
6. Least, B., S. Greci, M. Konopczynski and K. Thornton, “Inflow control devices improve production in heavy oil wells,” SPE paper doi:10.2118/167414-MS, Oct. 26, 2013.
7. Mathiesen, V, H. Aakre, B. Werswick, and G. Elseth, “The Autonomous RCP Valve – New Technology for Inflow Control in Horizontal Wells,” SPE paper 145737, SPE Offshore Europe Oil and Gas Conference and Exhibition, Aberdeen, UK, Sept. 6-8, 2011.
8. Halvorsen, M., and G. Elseth, “Increased oil production at Troll by autonomous inflow control with RCP valves,” SPE paper 159634, SPE Annual Technical Conference and Exhibition, San Antonio, Texas, Oct. 8-10, 2012
9. Halvorsen, M., M. Madsen, M. Vikøren Mo, I. Isma Mohd, and A. Green, “Enhanced Oil Recovery on Troll Field by Implementing Autonomous Inflow Control Device,”. SPE doi:10.2118/180037-MS, April 20, 2016.

_{ACKNOWLEDGMENTS}

^{An abridged copy of the paper, SPE-197700-MS, was presented at ADIPEC in Abu Dhabi, UAE, Nov. 11-14, 2019.}

About the Authors

Mojtaba Moradi

Tendeka

Mojtaba Moradi is a Subsurface manager at Tendeka in Aberdeen. He holds a PhD (2016) in petroleum engineering from Heriot-Watt University. He is a member of the European Association of Geoscientists and Engineers (EAGE), and SPE.

Michael Konopczynski

Tendeka

Michael Konopczynski is a director of subsurface engineering at Tendeka. He started his career with Shell Canada Limited in a variety of production engineering and technology roles for close to 20 years. His assignments with Shell included projects for steam-assisted thermal recovery, CO2 enhanced recovery, deep sour gas development, and gas-condensate developments in Canada, the United States, and the Sultanate of Oman. Following Shell, Mr. Konopczynski was part of the creation and growth of WellDynamics, where he served as Vice President of Technology, Product Line Management and Marketing. Following the acquisition of WellDynamics by Halliburton in 2008, he continued his role as director of Technology for Intelligent Completions and director of Reservoir Solutions. He is a member of the Society of Petroleum Engineers (SPE), is an SPE Short Course Instructor, and has authored numerous technical papers. He is based in Houston, Texas.