SPE-191635-MS

Production Performance of Multiple Completion Designs: Openhole, Slotted

Liner, ICD, and AICD: A Case Study for Water Control in Villano Field,
Ecuador

Alejandro Andrade, Mario Chango, Gustavo Atahualpa, and Ramón Correa, ENI; Georgina Corona,
Byron Calvopina, Juan Pico, Halliburton

Copyright 2018, Society of Petroleum Engineers

This paper was prepared for presentation at the 2018 SPE Annual Technical Conference and Exhibition held in Dallas, Texas, 24-26 September 2018.

This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents
of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written
consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may
not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract
This paper presents an analysis of the effectiveness of various downhole completion designs in reducing
or deferring water production in a mature field under the presence of an active strong aquifer reservoir.
The results of completions using nozzle inflow control devices (ICDs) and fluidic diode autonomous ICDs
(AICDs) are compared with conventional openhole and slotted liner completions. As all of these designs
were installed in the same field/reservoir, the comparisons provide a meaningful and representative analysis
of well production performance to assist in the identification of the most appropriate completion design
for future wells and the production optimization of existing ones. The designed vs. actual production
performance of inflow control completions providing water control (ICD/AICD) is presented and discussed.
The methodology was developed from comparative analysis results of conventional openhole and slotted
liner vs. ICD and AICD completions. The analysis was primarily based on elapsed time comparisons for
water/oil ratio (WOR) and water cut (WC) and used diagnostic plots to identify the water production
mechanisms, historical drawdown (DD), productivity index (PI), and production cumulative performance
(oil and water). The corrective actions are described, including operational procedures to reduce skin
damage and screen plugging implemented in the Villano-23HST2 (V-23HST2) well, which is the longest
horizontal well drilled in Ecuador and completed using AICDs; these corrective actions were based on
lessons learned from the Villano-22D (V-22D) well, which included appropriate fluid [brine/oil-based mud
(OBM)] conditioning, fluid displacement, filter cake removal, and compatibility testing between screens
and the fluid in which the bottomhole assembly (BHA) was deployed. Additionally, this paper evaluates the
importance of the design phase, emphasizing the importance of comparing preliminary data (permeability
and water saturation) compared to actual results obtained from the initial production test. Finally, as good
production results largely depend on successful operational procedures and execution, lessons learned and
best practices for deploying downhole completions in future operations for the Villano field in Ecuador are
discussed.
Although many studies compare ICDs vs. conventional completions, few compare different inflow
control technologies, such as ICDs vs. AICDs, within the same reservoir and with similar well conditions.
2 SPE-191635-MS

This paper compares various inflow control technologies in the same field with cumulative production data,
which verifies the effectiveness of each completion design. Based on these results, a validated methodology
for ICD and AICD simulations and design is also described as the basis for achieving good production
results.

Introduction
Water production is often a significant concern during the life of an oil and gas well. Increased WC
sometimes rapidly leads to significant increases in operating costs, resulting in uneconomical production
and possibly premature well abandonment and a loss of reserves. Almost all oil and gas reservoirs produce
water, and in mature or older fields, water represents a large percentage of total fluids production. In
this context, of particular interest are the adverse effects on production efficiency and future recoverable
reserves, the economics of water containment and disposal, and environmental impacts. Sometimes, the
main benefit of water-control treatments is simply the reduction of water disposal costs; however, excessive
water production is generally associated with a specific issue or combination of issues. The "water issue
types" are generally categorized according to water flow, which can be an open flow path, edge water from
waterflood or a natural aquifer, or bottomwater. These mechanisms are of interest because each has its own
typical water production history, which can help diagnose the issue. Various techniques can be used for
solving the issue of "excessive unwanted fluid production." Typically, these are divided into two different
categories: mechanical and chemical methods. As such, Villano field completion designs have evolved from
conventional openhole completions to the incorporation of ICDs/AICDs technologies to efficiently control
water production, with good results.

Field and Well Description

Eni, through its subsidiary Agip Oil Ecuador (AOE), is the operator of Block 10. The Villano field is situated
in a tropical rainforest area in the Pastaza Province in the east basin of Ecuador. The field was originally
discovered in 1950 by Shell Global and later by ARCO Oriente in 1992. Production began in May 1999,
and it has been operated 100% by AOE since 2000. The development was completed in six phases, which
included the construction of facilities, flowlines, and well pads and the drilling of production and water
disposal wells (WDWs). After 19 years of production, the total cumulative produced oil (19°API) is 145
million STB from 20 active production wells (18 from Hollin Sst and 2 from Napo T Sst) and five WDWs
[three in the Villano A well pad and two in the central processing facility (CPF)]. The current daily oil
production is approximately 12,900 BOPD associated with a water production volume of 204,000 BWPD
(which is 100% re-injected). The water disposal in the Villano and CPF occurs in the Hollin formation of the
Cretaceous age [below the oil-water contact (OWC)]. This formation has good characteristics in terms of
porosity (18%), permeability (0.5 md to 2 darcies), lateral continuity, and pressure support from an infinite-
acting aquifer that creates a strong water-drive production mechanism.
The majority of the wells are horizontal with 1,500 to 2,500 ft of openhole length. The oil/water mobility
ratio is 20 to 40, which means that water mobility is greater than that of oil, resulting in a significant amount
of water per well (10,000 to 30,000 BWPD). The oil production is constrained by the maximum fluid
handling capacity. Fig. 1 summarizes the Block 10 primary values and the average rock and fluid properties
in Villano's Hollín reservoir.
SPE-191635-MS 3

Figure 1—Block 10 location and field background.

Completion Design Considerations

The approach to completion design needs to be interdisciplinary, involving reservoir engineering, petroleum
engineering, production engineering, and drilling-completion engineering. The design process consists of
three phases: conceptual design, detailed design, and procurement. The primary function of a completion
is to produce hydrocarbons to surface or deliver injection fluids to formations; however, a completion
also needs to satisfy many other functions required for safety, optimizing production, servicing, pressure
monitoring, and reservoir maintenance. As described previously, most of the Villano field wells are
horizontal and the completion designs have been evolving from openhole and with slotted liners to water-
control solutions, such as ICDs and AICDs. The primary reservoir factors influencing completion design
that are to be considered include the production zone isolation distance from fluid contacts, secondary
targets, minimum zone separation, interval length, and future considerations. Fig. 2 shows a typical screens
configuration for a horizontal well.

Figure 2—Typical well configuration using screen technologies.

4 SPE-191635-MS

Conventional Completions
Openhole/Slotted Liner/Casing and Cemented (Perforated Completions)
In openhole sandface completions, the essential requirement is to maintain borehole stability during the
drilling and completion phase of well construction. Openhole sandface completions are usually avoided in
formations with several sand and shale sequences. A slotted liner is used where there is a risk of wellbore
instability to maintain a bore through the formation that otherwise might collapse and plug off all production.
It also assists in liquid lift because of the smaller flowing area. Slotted liners are oilfield tubular with small
milled slot openings and are manufactured in a variety of geometries. The single-slotted, staggered-row
pattern is generally preferred because a greater portion of the original strength of the pipe is retained. The
staggered pattern also provides a more uniform distribution of slots over the surface area of the pipe. The
performance is usually determined based on the open area presented to the formation. However, the pressure
loss through an open slot is much less than that which is caused by flow convergence in the permeable
media near the wellbore. Consequently, slot spacing is even more important because it controls the extent
of flow convergence away from the liner and into the formation.
Perforated completions are the most common globally because of the selectivity, flexibility, lower
costs, and convenience provided. Key issues with casedhole completion design include perforated interval
selection, gun type, shot density, underbalance or overbalance, and perforating method [i.e., casing guns,
through tubing guns, or tubing conveyed perforating (TCP)]. Additionally, the completion fluids program
is selected with regard to fluid quality and formation damage. It is also necessary to consider the formation
type, whether special perforating techniques are required (e.g., high shot density, ultradeep penetration, or
stimulation treatments), and achieving effective zonal isolation resulting from cement quality and distance
between zones. Fig. 3 describes typical conventional completions.

Figure 3—Typical conventional completions: slotted liner/perforated.

Water-Control Completions
Nozzle-Based ICDs
The viscosity differences between oil and water in the Villano field create an unfavorable mobility ratio;
the water flows at a higher velocity than the oil. Any method to delay and stabilize the water in the well can
assist in optimizing oil production. ICDs were evaluated for this purpose.
The short technology screening process concluded that the most appropriate technology was the "passive"
nozzle-based ICD. Other available technologies, which included channels or tubes, created pressure drops
by shearing the fluid, making the pressure drop dependent on the fluid viscosity and resulting in choking
back the oil. The pressure drop through the nozzle is a result of the static energy being converted to kinetic
SPE-191635-MS 5

energy and absorbed in the fluid downstream of the nozzle, as described by the Bernoulli equation. Nozzle-
based ICDs (Fig. 4) operate independently from viscosity.

Figure 4—Behavior/description of nozzle-based ICD installed in Villano (from Correa et al. 2015).

Influx differences from the reservoir can result in premature water or gas breakthrough, leaving valuable
reserves in the ground. ICDs are designed to improve completion performance and efficiency by balancing
inflow throughout the length of a completion. Placement of these ICDs is typically incorporated into
the design phase of a project and adjusted with final measurement-while-drilling (MWD)/logging-while-
drilling (LWD) information. The ICD is a nozzle-type configuration offering adjustability that can be preset
to a wide range of specific requirements before installation without the use of specialty tools.

AICDs
Horizontal and deviated wellbores provide access to narrow, oil-bearing formations for maximum contact
with pay zones. When the unbalanced influx from the reservoir causes water to migrate to the wellbore,
uneven production distribution is created. ICD settings can be adjusted for a target pressure decline and flow
rate and contribute to balancing the inflow. However, once downhole, the ICD settings for flow restriction
remain fixed. Fluidic-diode AICDs autonomously adjust the pressure drop and flow rate through the device
based on downhole fluid properties. These types of devices are more effective than ICDs in optimizing
oil production and for restricting water and gas production (Fripp et al. 2013). The utilized AICDs use
fluidic technology engineered as flow path channels to direct the fluid autonomously through a low- or
high-resistance path. They function similar to ICDs during oil production, while restricting the production
of water and gas upon breakthrough to minimize water and gas cuts. AICDs in Well Villano 23H have no
moving parts, require no downhole orientation, and adjust their flow rate based on the properties of the
downhole fluids. The fluidic-diode AICD has four different designs; Table 1 shows viscosity ranges and
applications. Based on downhole oil viscosity of Well Villano 23H, the selected device was AICD Range 3.

Table 1—Fluidic-diode oil viscosity range and restricted fluid.

The AICD contains two fluid dynamic components, a viscosity selector and a flow restrictor, both of
which function together to allow or restrict the flow of fluid without moving parts. The viscosity selector
uses a system of channels that, based primarily on fluid viscosity, "identifies" the fluid(s) that are flowing
through the AICD and then directs how the flow stream enters the flow restrictor. The higher viscosity
6 SPE-191635-MS

fluid (oil) takes a direct, radial, pathway to the exit, while the less viscous fluid (water and/or gas) takes
a tangential pathway to the exit. (Fig. 5). Based on the viscosity selector's output, the flow restrictor
significantly increases the restriction of the unwanted fluid (water and/or gas) while providing minimal
restriction to the production of the desired fluid (oil). The increased flow restriction for water is gradual
as the ratio of this fluid increases.

Figure 5—(Left) Oil flow path and (right) water/gas flow path fluidic-diode AICD.

Flow performance tests are used to measure the performance of the AICD at representative downhole
conditions. Fig. 6 shows single-flow performance testing through one single insert using Calpar 150 oil to
achieve 17 cp (122°F) and 36 cp (93°F) and water at 112°F for 0.6 cp viscosity. The graph illustrates that
the fluidic-diode AICD produces oil at a greater flow rate than water at the same differential pressure. Oil
at 17 and 36 cp is produced at similar flow rates through the device. However, when water flows through
the AICD, comparing at a 100-psi pressure differential, the flow rate is reduced approximately 48%.

Figure 6—AICD Range 3 (single insert)—oil and water flow performance.

SPE-191635-MS 7

The restriction of the fluidic-diode AICD is reversible; therefore, if water saturation recedes near the
wellbore, the flow restriction through the AICD will also decrease. The differential pressure across the
AICD decreases as the oil fraction increases, thus increasing oil production.

Villano Wells Completion Summary

Table 2 contains a completion summary of Villano wells. Different colors represent the wells used for the
production data analysis, 65% of which are horizontal. Water-control technologies have been implemented
since 2010.

Table 2—Villano well completion summary.

Production Data Analysis

This section describes the comparative analysis of conventional openhole and slotted liner vs. ICD and
AICD completions, in terms of WC %, WOR, PI, DD, and fluid rates. Horizontal and deviated wells were
separated to provide a better comparison.

Horizontal Wells
Because of the strong water drive mechanism, it can be assumed that water saturation into the reservoir
during the productive life of the field will increase. Fig. 7 shows the anticipated time to water breakthrough:
500 days for the wells drilled and completed from 1999 to 2000 and 10 days for the last well drilled in
2017. Additionally, the wells did not initially start with 0% WC; currently, it is common that the majority
of wells begin with a basic sediment and water (BS&W) greater than 60%. Even with such a scenario, the
plot shows that the water-control wells (in red) have a different trend compared with openhole and slotted-
liner wells (in blue). This means the mechanisms installed (ICD/AICD) are changing the normal production
behavior of this type of reservoir; even the Kv/Kh is almost ~1. Fig. 8 shows the WOR vs. cumulative
8 SPE-191635-MS

oil trend, which is the basis for reserves estimation. The water-control mechanisms are not detrimental to
reserves; in some cases, they maintain the reserves and other times the reserves increase with time. Fig. 9
shows the effect of water control on the well behavior. A clear difference can be observed in the PI and
DD of the openhole and slotted liner wells vs. the ICD/AICD wells. The water-control wells have, from
the onset, created an additional mechanical skin to the well. This "good skin" helps prevent and delay the
water influx. For this reason, these types of wells have less productivity and require more DD to produce
the desired fluid targets. In the Villano field, the openhole and slotted liner wells typically produce rates
greater than 20,000 to 30,000 BFPD. The water-control wells have maximum rate values of approximately
15,000 BFPD (Fig. 10). In most fields, water production represents a bottleneck. As such, the completion
strategy in the Villano field changed from conventional slotted liner completion to ICD/AICD. The lower
rates allow the capability to manage the actual total production rates in an efficient manner.

Figure 7—WC vs. normalized time.

Figure 8—WOR vs. cumulative oil.

SPE-191635-MS 9

Figure 9—PI vs. DD.

Figure 10—Fluid rate vs. normalized time.

Deviated Wells. As previously discussed, ICDs are used primarily to reduce the heel-toe effect in horizontal
wells, which delays water breakthrough; for deviated wells, the ICDs also delay early water breakthrough.
Additionally, it is important to consider that the ICD completion selection for a 66° deviated well is adequate
because the quality of the initial casing cementation is uncertain for a well with greater than 45° inclination.
Fig. 11 shows the BS&W vs. normalized time. The water-control wells (in red) show a better trend than the
analogous casing and cemented wells. Fig. 12 shows the effect on the reserves. Well V-18 ICD, with almost
3 million STB of production, has a better trend compared to the analogous Well V-22D AICD, which is not
performing as expected resulting from issues during the installation phase of the device. This emphasizes
that achieving production targets requires a well-defined procedure for running in hole (RIH) devices. After
analysis, a high skin (out of normal range) was detected in these wells, mostly related to screen plugging
in the upper AICD completion compartments.
10 SPE-191635-MS

Figure 11—WC vs. normalized time.

Figure 12—WOR vs. cumulative oil.

Fig. 13 shows the low PI for Well V-22D, which performed differently than the ICD and perforated
wells in terms of fluid production. Fig. 14 shows the production difference between a deviated well
with and without water-control mechanisms. Water-control wells were in the range of 6,000 BFPD while
conventional wells produced at rates from 12,000 to 24,000 BFPD.
SPE-191635-MS 11

Figure 13—PI vs. DD.

Figure 14—Fluid rate vs. normalized time.

ICD vs. AICD Performance

The first part of the analysis focused on horizontal wells, comparing the cumulative rates. While the AICD
(V-23HST2) producer well produced 50% (from ~400,000 B/D to ~600,000 B/D) to 500% (from ~100,000
B/D to ~600,000 B/D) more oil compared to at 300 days (black dot in Figs. 15 and 16 indicate V-23HST2) for
the ICD wells (Fig. 15), it also produced water because of the higher total fluid volume (Fig. 16). The higher
total PI might not necessarily represent the best producer because it combines oil and water production, but
it is needed to evaluate the WC% with regards to oil production. In this scenario, the best key performance
indicator (KPI) to use for comparing both technologies is to separate the PI into two components, oil PI and
water PI, and to plot a ratio between the BS&W vs. the oil PI. The well having the higher oil PI with less
WC would represent the best producer.
12 SPE-191635-MS

Figure 15—Cumulative oil AICD vs. ICD.

Figure 16—Cumulative water AICD vs. ICD.

Fig. 17 shows the ratio between BS&W vs. oil PI for the water-control wells (ICD/AICD). The same
analysis was performed for the deviated wells (Fig. 18). For the horizontal wells, the AICD wells performed
better than ICD completions. However, for the deviated wells, the ICD completions performed better, which
could be related to the high skin and poor operational practices, such as inadequate well cleanup and fluid
conditioning during RIH, leading to screen plugging in the upper section of the V-22D well.

Figure 17—Horizontal well BS&W vs. oil PI.

SPE-191635-MS 13

Figure 18—Deviated well BS&W vs. oil PI.

As described previously, the water-control mechanisms represent a good strategy for developing Villano
field wells. Placing special attention on the completion design is one of the most important factors to
achieving the oil-water targets.

Simulation Completion Design

Modeling fluid flow through complex completions is beyond traditional reservoir simulation and nodal
capabilities. Models should combine an accurate reservoir inflow with a highly detailed wellbore model. The
simulator used bridges the reservoir simulators and lift design software. It combines reservoir deliverability
with completion flow performance to enable rapid modeling of complex well hydraulics. The simulator
covers wellbore and near-wellbore (NWB) regions with a two-dimensional (2D) mesh of nodes (similar to
reservoir simulators), accounting for radius symmetry. Using this approach, a continuous, highly detailed
flow map between the reservoir, tubing, and up to four concentric annuli is possible. Instead of relying
on tank or PI models, as is typical in nodal applications, the simulator imports reservoir grid files to
simulate inflow. A software numerical solver generates pressures, flow directions, rates, void fractions, fluid
properties, temperature, inflow performance relationship (IPR), and PI.
The last drilling campaign, "Phase VI" performed during 2017, included three producers from the Villano
A pad (V-22D, V-23HST2, and V-24H), of which V-22D and V-23HST2 were completed using AICDs.
Before beginning the AICD design, it was necessary to obtain a previous analysis considering completion
tools, equipment available, and delivery time. The initial designs used the predrilling parameters and
synthetic logs.
During drilling, the logs were obtained in real time. This helped quickly obtain the necessary parameters
for the actual analysis and design. The simulation design began with 1) a numerical simulation model
containing the reservoir parameters in real time (static-dynamic properties), 2) well input parameters
obtained for drilling data, 3) petrophysical parameters (permeability and water saturation logs) obtained
from log interpretation, 4) the completion design, which was dependent on the available tools at location,
and 5) simulation results, such as pressure at the first node, pressure decline, oil rate, water rate, WC, and
PI. Fig. 19 shows the workflow for the design of a water-control (AICD) completion.
14 SPE-191635-MS

Figure 19—Workflow for design of an AICD completion.

AICD Completion Design

The workflow performed on the V-22D and V-23HST2 wells is discussed in the following sections. During
the simulation, one of the most sensitive parameters was permeability, so Kv/Kh was 0.8 to provide
connectivity between the reservoir layers of the model.

Input Data
Table 3 shows details of the well input parameters and petrophysical and pressure, volume, temperature
(PVT) properties.

Table 3—Well input parameters and petrophysical and PVT properties.

SPE-191635-MS 15

Analysis and Completion Design

The designs were based on delaying water breakthrough. Fig. 20 shows the V-22D well completion with a
total of 13 AICDs and 11 swellable packers (three water and eight oil) between the interval of 12,453 and
12,779 ft. Fig. 21 shows the V-23HST2 completion with a total of 20 AICDs and 13 swellable packers (five
water and eight oil) between the interval of 16,459 and 17,660 ft.

Figure 20—V-22D AICD completion.

Figure 21—V-23HST2 AICD completion.

Simulation Results
The simulations considered total liquid production of 6,000 BFPD for V-22D and 10,000 BFPD for
V-23HST2. Tables 4 and 5 show the main simulation results for V-22D and V-23HST2, respectively.

Table 4—V-22D simulations at expected initial conditions.

16 SPE-191635-MS

Table 5—V-23HST2 simulations at expected initial conditions.

Design vs. Initial

Well V-22D
This well began production in February 2017, with a daily peak rate of 1,276 BOPD. The performance was
less than expected because of operational issues experienced during the completion phase. The well was
completed using an electrical submersible pump (ESP) as the lift system.
The well was designed for a 6,000-BFPD maximum fluid rate. The expected initial WC% was 20% with
a PI of 3. The well began with 65%WC and a 1.8 PI. To achieve a match between simulation and actual
data, the water saturation and skin were modified according to the production test and buildup test. The
wells in the area are typically designed for higher fluid rates, so when WC% increases, the total fluid rate
is increased as well to increase oil production. Table 6 shows two cases (initial and actual conditions).

Table 6—Initial and actual conditions.

Fig. 22 shows the effect of using AICD completions. The blue line corresponds to design, the purple line
corresponds to the initial conditions, and the red line corresponds to actual conditions. According to the
analysis, the screens in the upper producer levels were partially plugged, resulting in a reduction in the PI.
High skin was evident during the buildup test. Additionally, oil production was affected because the well
was near OWC. These results indicate that the AICD completion did not perform as expected. The choke
was actually centered in the lower part.
SPE-191635-MS 17

Figure 22—V-22D AICD completion.

Well V-23HST2
This well began production in June 2017 and was completed using an ESP as the lifting system. The
performance exceeded expectations as illustrated in Fig. 15, with a daily peak oil rate of 3,058 BOPD. The
WC and fluid rate increased with time, as expected.
The well was designed for a 10,000-BFPD fluid rate. The expected initial WC% was 14%, and it started
with 50%WC. To achieve a match between simulation and actual data, the water saturation was modified
according to the production test. As mentioned previously, as the WC% increases, the fluid rates are
increased to gain more cumulative oil production, and the AICD design needs to account for these total
flow rates. Table 7 shows initial and actual conditions.

Table 7—Initial and actual conditions.

Fig. 23 shows the effect of using the AICD completion. The blue line corresponds to design, the purple
line corresponds to initial conditions, and the red line corresponds to actual conditions. These results indicate
that the AICD completion performed as expected with the higher initial WC%; the maximum choke was
located in the lower part, allowing equalization with the rest of the completion, maximizing oil production
and delaying the breakwater.
As such, it is strongly recommended to revise the input data before simulation because the software tends
to be optimistic in terms of WC.
18 SPE-191635-MS

Figure 23—V-23HST2 AICD completion.

Operational Procedure, Lessons Learned, and Best Practices

The well completion design is important for achieving the well and asset objectives. However, deploying the
completion to total depth (TD) should follow best practices for a successful completion operation, aligning
a multidisciplinary effort from the reservoir, fluids, and completion departments.
ICDs and AICDs are paired with a screen as a means of sand control or debris filter. Therefore, similar
practices as when running a standalone screen (SAS) completion should be followed. The following
minimum operational practices are recommended:

• Perform torque and drag (T&D) to predict the deployment feasibility of the completion string.

• Perform well cleanup after drilling and before RIH the completion. This step is important for
preparing the packer setting area as well as drifting the well for the completion string and removing
solids and residues to an acceptable level to allow completion installation.
• If OBM will be displaced by brine, perform compatibility tests to verify no emulsion is formed.

• Condition completion fluid before RIH completion.

• Effectively remove mud cake.

High skin (S ~20) was recorded in Well V-22D through a buildup test. A root cause analysis was
performed, analyzing each operational step. The main factor identified was mechanical skin associated
with solids not being removed—particularly in the last reservoir compartment, where challenges were
experienced while RIH—and plugging the screen, obstructing the flow through the screens. Most of the
operational practices previously discussed had been followed, but there was still opportunity to improve.
New best practices were developed and implemented in the next well, V-23HST2. The primary changes
included adding a magnet and junk basket to the well cleanup, extensive OBM conditioning, which included
production screen testing (PST) with screen coupons and increasing the screen gauge/slot from 330 to 787
microns to help prevent potential screen plugging, and a higher-density mud cake removal system to help
SPE-191635-MS 19

prevent fluid migration. The efficiency of these actions was confirmed with a reduction in skin, as detailed
in the following sections.

Clean Well Tools

Once the well was drilled, a clean well tool (CWT) BHA with stabilizers was run, which included the
following components:

• Until TD in the 6-in. openhole: 6-in. bit; 5 7/8-in. stabilizer; 47-ft, 4 3/4-in. drill collar (DC); and
5 7/8-in. stabilizer.
• Until 30 ft inside the 7-in. liner: scraper; junk basket with a fluid filter; magnet; brush; 3 1/2-in.
heavyweight drillpipe (HWDP); jar; and accelerator.
• Inside the 9 5/8-in. casing: scraper; junk basket with a fluid filter; magnet; and brush.

The primary purposes of this BHA were to

• Calibrate the openhole using a rigid BHA that included stabilizers to condition and simulate the
final completion BHA with swellable packers and AICDs tight to the openhole section
• Scrape the shoe track of the previous 7-in. liner to reduce the risk of a restriction in any cement
layer during running of the final openhole completion
• With the BHA on bottom, remove drilling OBM using a filter OBM to run the final completion

• Clean the fluid mechanically using a basket with a debris filter and magnet to remove ferrous
magnetic material

Mud Control
The OBM used provided a proper bridge to minimize formation damage and form an optimal filter cake.
During drilling operations, the mud composition was adjusted through software analysis and pore plugging
tests (PPTs).

Mud Conditioning
Because of the V-23HST2 wellbore geometry, the presence of coal layers with kaolinite-rich shale, and
high T&D estimated when running the openhole completion, the initial plan of using a recommended clean
brine was altered. T&D simulation showed the OBM lubricity would reduce T&D and ensure well stability,
which increased the likelihood of reaching final depth. Once the OBM was established as the running fluid,
actions were taken to reduce the likelihood of screen plugging resulting from the completion fluid. This
test is used to confirm that mud has been conditioned properly without showing retention of solids on the
screen coupon (Al-Ansari et al. 2017).
Mud conditioning is necessary to ensure that the solids contained within the mud do not exceed a certain
size (Fleming and Appleby. 2006). The solids can be derived from bridging agents or from drilled solids.
Bridging agents are particles added to the mud base fluid to primarily form a filter cake to help prevent fluid
loss to the formation by bridging in the formation pore spaces. Drilled solids are residue from the drilling
process. Conditioning the mud helps ensure minimal plugging of the screen completion during deployment
into the wellbore.
For Well V-23HST2, the OBM was conditioned at the shale shakers using API 170-mesh screen to remove
any particles or solids larger than 100 μm before pulling out of hole (POOH) with CWT. To verify the
effectiveness of the mud conditioning, a PST was performed at the rigsite using a screen coupon. Screen
coupons are round discs of filtration media. Their diameter is sized to match the PST test cell. The PST
test cell is an apparatus that holds the screen coupon. Low pressure, typically 10 to 20 psi, is applied to the
mud samples, and the time to pass through the coupon is measured, which should not be longer than 15 to
20 SPE-191635-MS

20 seconds. Fig. 24 shows the test confirmed that mud had been conditioned properly without indicating
solids retention on the screen coupon.

Figure 24—Screen coupon after PST, verifying mud is properly conditioned.

Other Actions
On completion of drilling the openhole section, the entire openhole should be backreamed to dislodge excess
filter cake (Fleming and Appleby 2006). Additionally, two hole volumes should be circulated at a suitable
rate to lift all solids completely out of the openhole, casing, and riser (if applicable). A solids-free system
can be introduced to the wellbore at this stage.

Equipment Considerations
To increase the probability of reaching TD, tools used in the final BHA had special features:

• Packer: an expandable liner hanger allowed rotation and high slackoff weight to pass through tight
spots.
• Swellable packers: a slip-on system allowed placing packers at planned depth without issues
associated with spacing out and not requiring additional actions to set.

Hydraulic Simulation and Pumping Fluids Scheduled for Effective Mud Cake Removal
Because the openhole completion included swellable packers, AICDs, and a liner hanger with a robust outer
dimension (OD) (>5.6 in.) that reduced the annular section to circulate, it was established that a maximum
pumping rate of 4 bbl/min would avoid packoff risks and high circulating pressures. Laboratory testing and
hydraulic simulation were performed to establish a proper fluid pumping schedule to help ensure mud cake
removal at 4 bbl/min. Table 8 and Fig. 25 show the sequence of laboratory tests performed on a mud sample
using a rheometer sleeve. The composition after each stage is included.

Table 8—Laboratory test for spacer design.

SPE-191635-MS 21

Figure 25—Digital images of laboratory testing for spacer design showing each stage as explained in Table 8.

According to the hydraulic simulation at 4 bbl/min, the following volumes of each fluid were established
to meet contact time and fluid removal requirements. The sweep pill volume was increased from the
recommended 50 to 350 bbl because the annular velocity between the 9 5/8-in. casing and drillpipe was
below the recommended value using the planned pumping rate. The adjusted swell pill volume removed
previous fluid effectively (Table 9, Fig. 26).

Table 9—Completion specialized software output above spacer design.

Fig. 26 shows simulated pressure considering the pumping schedule during displacement, with a
maximum circulation pressure of approximately 3,600 psi. Initially, the fluid (thinner pill) was pumped with
0 bbl. The additional volume shown corresponds to the 9.7-lbm/gal filter brine used to displace fluids.

Figure 26—Pump pressure behavior during displacement.

Filter Cake Removal System (FCRS)

To help ensure the effective removal of filter cake, a laboratory test was performed using OBM to form the
filter cake on an aloxite disc using the PPT procedure (Trujillo et al. 2009) (Fig. 27). The filter cake was
22 SPE-191635-MS

immersed into the FCRS formulation (Table 10) at BHT, and the filter cake dissolution was observed and
pH measured until total dissolution was achieved. Pre- and post-test weight was compared to establish the
dissolution efficiency as a percentage (Table 10).

Figure 27—(Left) Aloxite disc with filter cake and (right) post-FCRS use.

Table 10—FCRS design details.

Conclusions
The following conclusions are a result of this work:

• The water-control completion using AICD/ICD technology provided significant benefits during
the field development, allowing well production with a low water PI and high oil PI. Because the
fluid handling capability is a concern in the Villano field, this technology helped optimize energy
use by producing wells with low fluid rates and good oil performance.
• AICD completions performed better than ICD completions in horizontal wells; however,
performance in deviated wells was not as expected because of operational issues experienced
during the installation phase.
• Emphasizing the input data (permeability and water saturation) during the design phase is the most
important factor for achieving a match with the actual well conditions.
• Because good production results largely depend on successful operational procedures and
execution, the lessons learned and best practices described here for deploying downhole
completions should be beneficial for future operations in the Villano field in Ecuador.

Acknowledgments
The authors thank Eni, AOE, and Halliburton for permission to publish this work. Appreciation is extended
to all the professionals who participated in this project.

Terminology
μm = microns, 1/1000 millimeters
bbl/min = barrels per minute
SPE-191635-MS 23

°F = degrees Fahrenheit
AICD = autonomous inflow control device
AOE = Agip Oil Ecuador
BFPD = barrels of fluid per day
BHA = bottomhole assembly
BHP = bottomhole pressure
BHT = bottomhole temperature
BOPD = barrels of oil per day
BS&W = basic sediment and water value
BWPD = barrels of water per day
CC = casing and cemented
CPF = central process facilities
CWT = clean well tool
DD = drawdown
DC = drill collar
ESP = electrical submersible pump
FCRS = filter cake removal system
GOR = gas-oil ratio
HWDP = heavyweight drill pipe
ICD = inflow control device
KPI = key performance indicator
lbm/bbl = pounds per barrel
lbm/gal = pounds per gallon
LWD = logging while drilling
MD = measured depth
MMS = metal mesh screen
MWD = measurement while drilling
OBM = oil-based mud
OWC = oil-water contact
PI = productivity index
POOH = pull out of the hole
PPT = pore plugging test
PSD = particle size distribution
psi = pounds per square inch
PST = production screen test
PVT = pressure, volume, temperature
RIH = run in hole
SAS = standalone screen
SBM = synthetic-based mud
SP = saturation pressure
TCP = tubing conveyed perforating
TD = total depth (ft)
T&D = torque and drag
TVD = true vertical depth
v/v = concentration in volume/volume ratio
WBM = water-based mud
WC = water cut
24 SPE-191635-MS

WDW = water disposal well

WOR = water/oil ratio
WWS = wire wrap screen

References
Al-Ansari, A., Parra, C., Abahussain, A. et al 2017. Reservoir Drill-In Fluid Minimizes Fluid Invasion and Mitigates
Differential Stuck Pipe with Improved Production Test Results. Presented at the SPE Middle East Oil & Gas Show
and Conference, Manama, Kingdom of Bahrain, 6–9 March. SPE-183764-MS. https://doi.org/10.2118/183764-MS.
Correa, R., Andrade, A., and Ippoliti, M. 2015. Well Production Performance in a Deviated ICD Completion: Case
Study with Analogous Cased and Perforated Wells. Villano Field – Ecuador. Presented at the SPE Latin American
and Caribbean Petroleum Engineering Conference, Quito, Ecuador, 18–20 November. SPE-177218-MS. https://
doi.org/10.2118/177218-MS.
Fleming, A.J.A. and Appleby, R. 2006. Wellbore Clean up Best Practices: A North Sea Operator's Experience. Presented at
the SPE/IADC Indian Drilling Technology Conference and Exhibition, Mumbai, India, 16–18 October. SPE-101967-
MS. https://doi.org/10.2118/101967-MS.
Fripp, M., Zhao, L., and Least, B. 2013. The Theory of a Fluidic Diode Autonomous Inflow Control Device. Presented at
the SPE Middle East Intelligent Energy Conference and Exhibition, Dubai, UAE, 28–30 October. SPE-167465-MS.
https://doi.org/10.2118/167465-MS.
Trujillo, H., Rueda, H., Prent, L. et al 2009. Delayed-Reaction Filter Cake Breaker Helped Reduce Formation Skin
and Increased Productivity Index in Multiple Horizontal Wells. Presented at the National Technical Conference &
Exhibition, New Orleans, Louisiana. AADE NTCE-08-05.