RESERVOIR MANAGEMENT & IMPROVED RECOVERY

Farhad Ebadi, Mark Reynolds, Dr. David Davies, Prof. Patrick W. M. Corbett
Institute Of Petroleum Engineering
Heriot-Watt University, Edinburgh, UK

ABSTRACT
Summary
Intelligent completions are focused on the delivery and management of increased
production flexibility. Intelligent Well systems Technology (IWsT) delivers the
ability to install, operate, monitor and control completions without the need for
conventional interventions. Multi-zone intelligent-well completions contain
appropriate monitoring devices located between zonal isolation packers. They
control the flow into or out of each zone with Interval Control Valves (ICV5).
Installation of intelligence at the well level is, unfortunately, not always a
guarantee for success - it must add value to gain acceptance. This paper starts
from the premise that some reservoir types are inherently more suited to IWsT
application than others.
A reservoir model should predict the hydrocarbon distribution and the reservoir
flow properties. The internal-makeup of a reservoir provides a framework of
connectivity and fluid flow pathways through the reservoir. The spatial distribution
of flow units and barriers, such as faults, can be represented within a reservoir
model. A probabilistic reservoir model tries to capture reality by logical relations;
but is limited by our perception, knowledge and understanding of reality. It is,
therefore, the best-guess.
A series of generic reservoir types based on property distributions derived from
field data, have been built. They were tested to determine the added value from
IWsT compared to standard well completions. Situations in which IWsT proved
particularly successful have been identified. Results show that IWsT can control
uneven, invading fluid-fronts that develop along the length of the wellbore due to
permeability differences, reservoir compartmentalisation or different strength
aquifer/gas cap support. Recovery improves and water production reduces with
the correct choice of ICVs (Downhole Interval Control Valves) number, position
and length of zone being controlled. However, the degree of improvement is
dependent on reservoir type and differs from one reservoir model to another.
Results from this study have helped to develop a selection criterion for improved
implementation of IWsT.
Introduction
An intelligent well is a well with the ability to control the production flow by a
down-hole choke. This is managed through real time monitoring and control of
the producing zones using installed Inflow Control Valves (ICV) and an optimised
sensor distribution for data acquisition and down-hole fluid production
measurement. It also has the ability to shut off water/gas producing zone at the
wellbore. It produces single or multiple zones into one wellbore, leading to
commingled production from different zones and lateral bores.
Intelligent Completions, by commingling stacked pay, manages the (possibly
tilted) oil rims with different thickness in multiple fault blocks from a single well

bore. Figure 1 illustrates the commingled production from separate sands using
an intelligent well.

Intelligent completion techniques are gaining popularity because of their reservoir

monitoring and well / field performance management capabilities while
minimizing the well intervention requirements. IWsT can decrease the risk and
uncertainty associated with the production of complicated reservoirs. Reservoir
management objectives include increasing production and reserves, maximising
recovery, minimising capital and operating costs while reducing risk. It involves
taking decisions and predicting their future consequences hence the need to
model the reservoirs future behaviour. This requires skills in reservoir
characterisation, reservoir performance, well performance and field development.
A key issue in the reservoir management process is geological uncertainty.
IWsT, operating at or near real-time, enables high quality, well surveillance,
interpretation and reaction in a continuous feedback loop. The ultimate goal for
this continuous monitoring of the reservoir is to implement a more proactive style
of reservoir management technique. Identifying the type of reservoir where the
IWsT can be applied is an important step in evaluating its application. Knowing
where to apply this technology begins with reservoir characterization- a crucial
stage in reservoir management.
A useful tool for this stage is the classification of reservoirs. Weber and Geuns
proposed that reservoirs could be grouped into three basic reservoir types:
Layer Cake Reservoir type, Jigsaw Puzzle Reservoir type and Labyrinth
Reservoir type. Deterministic Petrophysical Modeling uses interpolation to assign
a property value to cells that are not penetrated by wells. It is based on upscaled
well log data Stochastic Petrophysical Modeling generates multiple equi-probable
realizations of the reservoir, where there is heterogeneity below the facies scale.
Stochastic Modeling is a helpful tool for modeling the uncertain aspects of a
reservoir. It can produce a scale of multiple realisations that all fit the basic data
and field information.
Methodology
A range of reservoir models have been studied. These include:
1. A constant permeability, single reservoir
2. Multiple, stochastic realisations of a heterogeneous, single reservoir with no
faulting or reservoir dip
3. A faulted Reservoir with transmissive and sealing faults and a range of OilWater contacts and Reservoir Pressures

4. Multiple deterministic and stochastic realisations of heterogeneous inclined

and layered reservoirs with a range of dip angles
5. A channel sand model
The production from long horizontal well(s) modeled in the above reservoir types
was simulated using the multi-segment option in Eclipse Simulator. Figure 2
illustrates the well configuration modeled in Eclipse.

The optimization objective in different scenarios was to maximize oil recovery

while minimizing water production by managing the ICy setting. The optimization
policy was kept simple so that it was possible to compare the different scenarios.
Results
Results from this study showed that I-Well gives significantly better recovery in
most cases. The % recovery and the scale of the increase vary from a
reservoir model to another and is a function of the distribution of porosity and
permeability in the models. Figure 3a shows a simple flat model with inclined
layers. A horizontal well is placed near the top of the model. A variety of model
configurations have been used to illustrate the versatility of IWsT e.g. increasing
levels of cross-flow & communication have been achieved between the layers in
Figure 3b by progressively removing the shale barriers.

Figure 5 shows extra recovery with IWsT from the high and permeability layers compared
to the conventional completion shown 4. Figure 6 illustrates the reduction in water
production using intelligent The full results will be discussed in the paper and
presentation.

Conclusions
IWsT has been shown to be capable of managing geological variability and thus
coping with geological uncertainty in a wide range of reservoirs. IWsT can control
uneven, invading fluid-fronts that develop along the length of the wellbore due to
permeability differences, reservoir compartmentalisation or different strength
aquifer/gas cap support.
In the full paper, the applicability of Intelligent Wells has been classified and
discussed as a function of the reservoir type.
References
1. Weber K.J. and Van Geuns L.C., Framework for Constructing Clastic
Reservoir Simulation Models, JPT, 42(10), 1248-1253, 1296-1297