American Oil and Gas Reporter - November 2016 - 40

SpecialReport: Oil & Gas Computing
greater than 70 percent of the stage's
fluid volume and proppant concentrated
in the two perforation clusters closest to
the heel of the stage. By comparison,
there were no stages in which a similar
70 percent of the stage fluid volume and
proppant were concentrated in the perforation clusters closest to the toe.
In addition to heel bias, single perforation/fracture dominance was observed
consistently. Four of the 14 stages showed
a single perforation cluster taking two

times or more fluid volume and proppant
than the perforation cluster that took the
next largest fluid/proppant amount for
that stage. Two of the stages (stages 4
and 13) showed a single perforation
cluster taking three times more fluid/proppant than the perforation cluster that took
the next largest amounts.
Figure 3 is a DAS distribution profile
summary of proppant placement, showing
the breakdown of the total proppant that
exited any given cluster for all 14 stages.

FIGURE 3
Summary of DAS Distribution Profile of Proppant Placement
300,000

250,000

Proppant (lbs)

200,000

150,000

100,000

50,000

0
1

2

3

4

5

6

7

8

9

10

11

12

13

14

FIGURE 4
DTS (Top) and DAS (Middle) Profiles of Stage 4 Plotted in Time

Stage 4 Imported Pump Data
60

14,000

55

3

50

12,000

45
10,000

40
35

8,000

2

30
25

6,000

20
4,000

1

15
10

2,000

5
0

0
11:30:00 PM

12:00:00 AM

12:30:00 AM

1:00:00 AM

1:30:00 AM

40 THE AMERICAN OIL & GAS REPORTER

2:00:00 AM

2:30:00 AM

3:00:00 AM

3:30:00 AM

4:00:00 AM

0

Perforation inactivity was frequently observed as well. In nine of the 14 stages
(11 of the total 64 perforation clusters),
one or more perforation clusters took
less than 25 percent of the planned proppant.
In general, diverter effectiveness was
inconsistent. All 14 stages were pumped
with two or three sand ramps (the design
called for three each stage), with diverter
applied between the sand ramps.
Despite fairly similar surface pressure
responses attributed to the diverter, DAS
observations indicated that the downhole
diverter effect was quite different than
expected. The completion design assumed
the diverter would reduce or stop treatment
entry into any clusters taking a preponderance of the fracturing treatment, and
then divert the treatment into clusters
that had not received significant treatment.
As an example of this behavior, Figure
4 is a composite figure showing the DTS
(top), DAS (middle), and pump curve
(bottom) for stage 4. This stage consisted
of four perforation clusters, which are
shown to the left of the DAS/DTS traces.
The x-axis represents the time scale for
the treatment. Both the DTS and DAS
traces indicate treatment distribution into
all four clusters during the first ramp,
with a stronger signature in the heelmost clusters (top of the figure).
The diverter then was added during
the flush of this ramp. At approximately
the time this first diverter stage arrived at
the perforation clusters, the cluster closest
to the toe (bottom in the figure) was effectively shut off.
After the second diverter stage arrived
down hole (pumped during the flush of
the second stage), minimal change occurred in the treatment diversion. An
analysis of the calculated percentage of
proppant distribution associated with the
fiber behavior in Figure 4 makes it obvious
that the initial dominant cluster remained
the dominant cluster throughout the entire
treatment.
Fracture Modeling
The disparate allocation of treatment
volume among respective perforation
clusters within each frac stage raises the
critical question of what fracture geometry
is likely to develop from any given cluster.
Fracture modeling was undertaken to develop insights into fracture geometries,
proppant concentrations, and fracture
conductivities created by the completion.
American Oil and Gas Reporter - November 2016

Table of Contents for the Digital Edition of American Oil and Gas Reporter - November 2016
