Title : The Impact of Different Completion Parameters fractured horizontal wells is expected to create a

on Well Performance of Gas Unconventional complex sequence of flow regimes.

Reservoirs.
Authors: Well performance of gas unconventional reservoirs

Engineer: Ibrahim Salahudin Mohamed . is greatly affected by the different completion

configurations between the reservoir and the
Prof. Dr. Hamid M. Khattab.
horizontal well. Completion parameters that greatly
Assist. Prof. Dr. Ahmed A. Gawish.
affect the well performance include the number of
Dr. Mazher H. Ibrahim.
frac stages, frac face skin, production under
Abstract
different draw down or ( - ) and the effect of
Unconventional reservoirs are defined as reservoirs that
production from different drainage areas. These
cannot be produced at economic flow rates or that do not
parameters should be studied and quantified for
produce economic volumes of oil and gas without
proper analysis of long-term linear flow periods
assistance from massive stimulation treatments or special
associated with tight/shale gas production.
recovery process. Shale and tight reservoirs are low quality
reservoirs with low permeability. This low permeability Methodology

can be as low as 100 Nano decries. . Well performance in GASSIM simulator is used, GASSIM is a single phase, 2D
gas unconventional reservoirs is greatly affected by simulator that is originally developed to simulate real gas
different completion parameters of the horizontal wells. flow in both x-y and r-z domain. Later, it was modified to
These completion parameters include number of frac include the ability to simulate liquid case. It is written in
stages, frac face skin, production under different draw Visual Basic code for Excel. It has been used extensively
down or ( - ) and the effect of production from in previous studies and proved to give accurate solutions.

different drainage areas. In this study, these parameters

will be studied and quantified for proper analysis of long-
term linear flow periods associated with tight/shale gas
1. The effect of number of stages of the
production. hydraulic fractures.
Introduction The effect of the number of stages of the hydraulic
Horizontal drilling and multistage hydraulic fracture fractures on well performance of gas
are the two key enabling technologies for the unconventional reservoirs can be studied in terms of
economic development of ultralow permeability the frac spacing.
reservoirs as tight and shale gas reservoirs
From Fig.1 The spacing between two successive
(permeability in Nano Darcy).The use of multi-
fracs
= .……...….…..………………………………….. Stimulated volume for each stage is: .
Eq.1 ….…………………………………….……………
……….Eq.2
Where the horizontal well length and n is is the
number of stages. Eq.2 is valid if there is no drainage at the symmetry
line between to successive fracs, so that each stage
has its own drainage volume or in other words no
frac interference occurs.

Increasing the number of stages also

decrease the total time needed to deplete the
reservoir due to increasing the production rate as
shown in Fig.4 Mathematically:

The total production of the total stages = n*

Fig.1- Horizontal well with length and number of
stages n. production of single
For fixed well length, increasing the number of stage…………………………...….Eq.3
stages, decreases the frac spacing. The input data to Eq.3 is valid if we assume that all stages are
GASSIM is summarized in Table1. The production identically and equally spaced.
rates for different stages (n= 2, 5, 10, 15,20,25, 30 Although the initial rate increases with increasing
and 40) are shown in Fig. 2 The time to the end of the number of stages ,the ratio of the initial rate to
formation linear flow or the time to the end of half the rate at the beginning the boundary dominated
slope period ( ), estimated ultimate recovery flow decreases which indicates that the decline rate
(EUR) if the economic limit is 3MSCF/D, the is higher when more stages are added as shown in
recovery factor and the productivity index is Fig. 5.
summarized in Table 2. Increasing the number of stages also results in an
Increasing the number of the frac stages will result increase in the productivity index (PI).The increase
in a decrease in the time to the end of transient in the productivity index is a result of the increase
linear regime or a decrease in the time to the end of in the production rate with increasing the number of
half slope period as shown in Fig. 3 The time stages under constant bottom hole flowing pressure
to the end of half slope period is decreased due to i.e. constant as shown in Fig. 6.
the decrease in the stimulated drainage volume for
each stage which will cause a rapid depletion of the
reservoir fluid. Mathematically:
 Table 2- Summary of , EUR, total time, RF and
PI for different values of n.

Table 1- The input data to GASSIM to simulate the

effect of changing the number of stages on well
performance.

Fig. 3-The effect of increasing frac stages on the

transient linear period.

Fig. 2- The production rate for different stages on the

same graph .

Fig. 4-The effect of inceasing frac stages on the

total depletion time till economic limit is reached
.
 Fig. 5-The ratio between the initial rate to the Fig. 7–The cumulative production with the time for
rate at the beginning of the boundary dominated different stages.
flow. .

Fig. 6-The effect of increasing frac stages on the

productivity index.
Fig. 4. 8– The cumulative production different
Increasing the number of stages, increases stages after 14000 day.
the cumulative production as shown in Fig. 7 and
Fig. 4.8 . Fig. 4.9 is the percentage of production
increase for n> 2 and n>15 after 14000 day of
production . As shown in Fig.4.9, increasing the
number of stages from n=2 to n=5,10,15 will result
in a percentage of production increase of 60%,
73.6%, 75.5% while increasing the number of
stages from n=2 to n=40 results in a percentage of
production increased of 77%.Increasing the number
of stages from n=15 to n=40 results in percentage of
Fig. 4. 9- The percentage of production increase for
production increase of 5%. Similar results after
n> 2 and n>15 after 14000 day of production.
2000 day of production is shown in Fig.4.10 .
 2. Effect of fracture face skin.
Fracture face skin is modeled in two different
models as chocked fracture damage and fracture
face damage Chocked fracture damage occurs
when the region around the well bore has a reduced
permeability. Choked fracture damage originates
when the proppant within the fracture is embedded
or crushed or lost as shown in Fig.12 Where
Fig. 4. 10- The percentage of production increase for
, , are the chocked fracture length , the
n> 2 and n>15 after 2000 day of production.
permeability of the damaged zone and the width of
the damaged zone respectively. Fracture face
The number of stages should be economically
damage occurs due to the filtration of the fracturing
optimum and determined in terms of the in situe
fluid into the formation and formation of filter cake.
conditions of the reservoir , the estimated ultimate
Filtration and filter cake reduce the permeability in
recovery and the expected production period. Fig.
the region surrounding the frac as shown in Fig.4.13
11 is the percentage of production increase if the
. Different values of fracture face skin (0, 0.01,
production period extended from 2000 to 14000 day
0.02, 0.05, 0.07, 0.1, 0.2, 0.5, 0.7, 1) are used to
.As shown from Fig.11 that for n=2,5,10,15
study the impact of skin on the well performance.
extending the production period from 2000 to
The data used for simulation is summarized in
14000 will result in a percentage of production
Table 3. As the fracture face skin increases, the
increase of 60% ,58%,48%,35%while for n= 40
linear flow regimes dimensioned gradually and look
increasing the production period will result in
like another flow regimes. This is could lead to
average production increase of 3.8% .
wrong analysis for the well performance and
reservoir parameters estimated from wrong flow
regime as shown in Fig. 14.

Fig. 11- Percentage of production increase if the

production period extended from 2000 to 14000 day. Fig. 12- Choked fracture damage from
.
 Fig. 13- Hydraulic fracture with fracture face
damage from
.

For small values of frac face skin i.e ( =0.0.

=0.01, =0.05, =0.1), the linear flow
regime or half slope on log- log plot is changed at
Table 3- The data used for simulating the effect of
early time and the curves become more flatten but Skin on the well performance of gas unconventional
finally linear flow is reached or the half slope reservoirs.

appears .The flatness of the curves may cause

misinterpretation of the well performance during [ ]
The straight line specialized plot of
the early production period. The flat part of the
curve indicates a transient radial flow which is vs √ for different values of skin is shown in Fig.15
completely incorrect. Other note is that, although . As shown in Fig.15 that the linear flow behavior
the small values of skin change the shape of the still preserves its straight line shape which is a clear
production rate on the log-log plot, the total evident on the linear flow behavior but the effect of
production time is nearly the same i.e. the curves skin appears only as an intercept on y-axis
are matched. For large values of skin ( =0.5 .Increasing the skin will cause an increase in the
intercept value. The straight lines representing
, =1), the linear flow regime doesn't appear on
different values of skin are parallel to each other but
the log –log plot at all but complete transient radial
with different intercepts.
and boundary flow appear on the log-log plot and
the total production time increased .
 √ is plotted. The cumulative production for
different values of frac face skin is shown in Fig.16
.The percentage of production loss for different
values of skin is shown in Fig.17 In this case the
impact of frac face skin is 7.6%, 14%, 44% and
60% production loss for frac face skin of
0.05,0.1,0.5 and 1, respectively.

Fig. 14- The production rate for different skin

values.

Fig. 16–The cumulative production [Link] for

different values of frac face skin.

[ ]
Fig. 15-Plot of vs √ for different
values of skin.
From the above , the frac face skin causes
dramatic change on the log –log plot by masking
the actual linear behavior to be like another flow
regime .The linear flow still preserve its
characteristic shape on the specialized plot of
[ ]
vs √ as straight line so that

interpretation of the actual production data may be

completely incorrect if the analysis is limited to
[ ]
log-log plot till specialized plot of vs Fig. 17- Percentage of production loss for
different values of frac face skin.
 Fig. 18 . As it is shown in Fig. 18 that all the

3. Effect of Production under previous cases don't have the same slope as
obtained from the analytical solution. This indicate
different drawdown .
that linear flow analysis is greatly affected by the
In this section, we will investigate the effect of
drawdown.
production under different draw down on well
performance of gas unconventional reservoirs. Two To demonstrate this effect Ibrahim and
scenarios will be used to indicate the effect of the Wattenberger (2005 defined a dimensionless
draw down on well performance of gas draw down parameter defined as :
unconventional reservoirs. The first scenario used is
[ ]
performing the simulation cases under constant ...........................................Eq.6

initial reservoir pressure while bottom hole flowing

Then defined correction factor for correcting the
pressure changes for each case. During the
slope of straight line as:
production life of the reservoir ,the well may be
shut down for several reasons so the second =1-0.0852 -

scenario is performing the simulation cases with 0.0857 ………………………..………….Eq.7

different initial reservoir pressure while is To indicate the importance of the correction factor
constant. Finally performing the simulation cases [ ]
defined in Eq. 7 a plot of vs. √
under constant draw down but with different initial
reservoir pressure and bottom hole flowing pressure is shown in Fig.19 .The correction factor normalize

i.e. ( ) is constant while and aren't the different slopes straight lines shown in Fig. 18
into only one straight line with constant slope in
constants .
Fig. 19 Linear analysis without using the correction
3.1 Production under different draw down factor will result in an over estimation of the
while the initial reservoir pressure is
constant. calculated reservoir parameters as fracture half-
length , √ and OGIP.

In this case, different values of for each

simulation case is used while initial reservoir
pressure is constant. The data used for simulation is
summarized in Table 4.

[ ]
The straight line specialized plot of

vs.√ for the previous simulation cases is shown in

 [ ]
Fig. 19- vs √ for the different
values of at constant 5000 psi.

3.2 Production under constant while the

initial reservoir pressure isn't constant .

Different scenario to indicate the effect of

Table 3- The data used for simulating the effect
production under different draw down while the
of production under different on the well
initial reservoir pressure is not constant but is
performance through gas unconventional reservoirs.
constant. During the production period the well may
be shut for different reasons but when the
production is continued the reservoir pressure may
be not at the initial reservoir pressure during the
initial production period. To show the effect of
production under different initial reservoir pressure
different simulation cases are run with constant
. The input data used to run these simulation
cases is the same data in Table 4 but with
=2000 psi and =6000,5000,4000,3000 and
[ ]
Fig. 18- Plot of vs √ for the different 2500 psi.
values of at constant 5000 psi.
[ ]
A plot of vs. √ is shown in Fig. 20.

As shown in Fig. 20 that different straight lines

with different slopes is obtained for different .The line analysis to obtain the different reservoir and
reason for these different slopes straight lines is the frac properties.
reservoir and fluid pressure dependent properities

√( ) . To obtain a single straight line with

constant slope it is preffered to plot

[ ]
√ or
√( )

[ ]
√( ) √ to eleminate the

effect of the pressure dependent properities and

obtaining a single straight line with constant slope
as shown in Fig.21.
[ ]
Fig. 21-Plot of √( ) √ for
the different values of at constant 2000 psi.

4. The effect of Production from

different drainage area on well
performance
In this section we will study the effect of
production from different drainage areas on well
performance of gas unconventional reservoirs . This
effect will be studied by assuming that the
horizontal well isn’t completely penetrate the
reservoir .Assuming different horizontal well
Fig. 4. 20-Plot of vs √ for the
different values of at constant 2000 psi. lengthes ( with constant number of [Link] the
previous study of the impact of different fracture
stages we assumed a constant reservoir length but
The slope of the straight line in Fig.4.31 is
the number of stages was variable. The data used
which is constant for the same reservoir for simulationg the effect of different drainage areas
√

because it doesn't include any pressure dependent is summerised in Table [Link] well length is

properties. This slope can be used for the straight changed in each simulation case ( =1000, 2000,
3000, 4000 and 5000 ft).
 Fig. 22 - Log log plot of the previous simulated cases
under different drainage areas .

Table 4- The data used for simulating the effect of

production from different on the
well performance of gas unconventional reservoirs.

Table 5 -The time to the end of transient linear

The effect of different drainage areas or frac
for each case.
spacing on the well performance can be illustrated
by plotting all the previous cases on the same log
log plot as shown in Fig. 22.

As shown in Fig. 4.22, the plots for the different

cases overlaps each other until the transient linear
period is finished. Since the reservoir properties and
the fracture half-length are the same in all these
simulated cases so that they have the same slope on
the straight line specialized plot of
[ ] Fig. 4. 23- The time to the end of transient linear flow
√ . Table 6 and Fig. 23
period for each case.
summarize the calculated time to the end of
transient linear for each case.
 estimation of the reservoir properties.

Conclusions. 7. The length of the horizontal well throught

The major conclusions of this work can be out the reservoir only results in an increase

summarized as follows: in the time of the end of transient linear flow

but has no effect on the straight line
1. Well performance of gas un conventional
specialized plot .
reservoirs is greatly affected by the
completion configuration between the
horizontal well and reservoir.
NOMENCLATURE
= half width of the reservoir, ft
2. Increasing the number of stages not only
causes a rapid depletion of the reservoir but = frac half length ft

also result in a decrease in the time to the

= half distance between two successive fracs ,ft
end of half slope period.
= dimensionless pressure
3. A decrease in the time to the end of transient
linear regime makes performing (RTA) = dimensionless time based on half width of the
analysis difficult to be performed. reservoir
4. The number of stages should be
=dimensionless time based on half distance
economically optimum and determined in
between two successive fracs
terms of the in situe conditions of the
reservoir , the estimated ultimate recovery = dimensionless rate
and the expected production period
= dimensionless pseudo pressure
5. Interpretation of the actual production data
= pseudopressure at initial reservoir pressure,
may be completely incorrect if the analysis
is limited to log-log plot as the fracture face
skin masks the actual linear behavior to be pseudopressure at bottom hole flowing
like another flow regimes so the straight pressure ,
line specialized plot is required for proper
= gas flow rate, MSCF/D
interpretation as the skin appears only as an
intercept on y-axis . = reservoir tempreture, ,

6. RTA is a draw down dependent analysis so = reservoir permeability , md

that performing RTA without the use of the
= total compressibility,
correction factor developed by Ibrahim and
wattenberger would result in over = gas viscosity, cp
 = porosity

=slope in case of constant rate

=slope in case of constant pressure

= reservoir thickness , ft

= cross sectional area to the flow,

= time to the end of half slope period ,Day

= pore volume

= original gas in place,BSCF.

= gas formation volume factor at initial

reservoir pressure ,RB/STB

= reservoir extension, ft

REFERENCES
(1) Cinco-Ley, H., Samaniego-V., F. 1981.
Transient Pressure Analysis: Finite
Conductivity Fracture Case Versus
Damaged Fracture Case. Paper SPE
10179 presented at the SPE Annual
Technical Conference and Exhibition,
San Antonio, Texas, USA, 4–7 October.
(2) Ibrahim, M. and Wattenbarger R.A.
2005. Rate Dependence of Transient
Linear Flow inTight Gas Wells. Paper
CIPC 2005-057 presented at Canadian
International
Petroleum Conference, Calgary, Alberta, 7–
9 June. DOI: 10.2118/2005-057.