See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/301397096

An Approach To Account ESP Head Degradation In Gassy Wells For ESP

Frequency Optimization

Conference Paper · October 2014

DOI: 10.2118/171338-MS

CITATIONS READS
2 198

3 authors, including:

K.V. Litvinenko Rinat Khabibullin

BashNIPIneft, LLC, Ufa, Russia Gubkin Russian State University of Oil and Gas
21 PUBLICATIONS 22 CITATIONS 67 PUBLICATIONS 143 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

ESP data-based prediction models via ML View project

Data Science in Oil & Gas View project

All content following this page was uploaded by Rinat Khabibullin on 11 February 2021.

The user has requested enhancement of the downloaded file.

SPE 171338

An approach to account ESP head degradation in gassy well for ESP

frequency optimization
Krasnov V.A. (Rosneft), Litvinenko K.V. (BashNIPIneft), Khabibullin R.A. (RSU of oil and gas)

Copyright 2014, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Artificial Lift Conference and Exhibition - North America held in Houston, Texas, USA, 6–8 October 2014.

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 necessar ily 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

The majority of Western Siberia oil wells in Russia equipped by electric submersible pumps (ESP).
Nowadays the common practice is to operate oil well with ESP as close to its production potential as possible.
This strategy allows achieving maximum oil rates and economic efficiency according to Vogel’s inflow
relationship.

Main limiting factor to decrease BHP and increase well’s oil production is an ability of ESP to work with
high free gas fraction at pump intake. It is well known that ESP fails to generate pressure drop with high gas
fraction at pump intake. However, there is no appropriate mathematical model to account it in ESP design
properly.

It is often assumed that ESP works well with gas fraction at pump intake below critical and cannot operate
with gas fraction at pump intake above critical. Critical gas fraction varies for different ESP types and
manufactures from 25% to 35% and even 50% for ESP with gas handling devices like MVP from Centrilift, AGH or
Poseidon from Schlumberger. Using this assumption one can conclude that affinity law should work well for ESP
with gas fraction below critical. Nevertheless, it is often observed on wells with high gas fraction that increasing
of ESP rotating frequency does not affect BHP and production as it predicted by affinity law.

This paper discusses an approach to describe an abnormal well behavior after frequency change. It shows
that correlation for ESP head degradation as in SPE 117414 combined with nodal analysis technique allows to
predict well behavior with high free gas fraction. It also allows optimizing energy consumption by ESP by varying
ESP rotating frequency. Field cases discussed.
2 SPE 171338

Nodal analysis overview

Nodal analysis introduced by Mach, Proano and Brown [Brown 1984] and described in detail in
[Brown 1984] is a powerful tool for oil well performance analysis. It is widely used for self-flowing well and
artificially lifted wells analysis and design. However, nodal analysis usefulness in not fully utilized yet. One of the
nodal analysis advantages – is its ability to present ESP and well performance characteristics in graphical form.
Charts and plots allows analyzing solution obtained and its sensitivity to initial data. Often it is useful to present
same data using different plots highlighting different aspects.

Oil well with ESP can be described using different tools. Most widely used tool is a well system plot used
for ESP design. Well system plot combines pump performance curves and well system curve (Figure 1). Well
system curve show difference in the outflow head and inflow head.

Figure 1 Well system curve example (plot from pump design software)

Well system plot is convenient for selecting pump type, number of stages and frequency because it
utilize pump performance curves provided by pump manufacturers.
SPE 171338 3

Other tool is nodal analysis plot for ESP well. Placing analysis node at bottom hole we can separately
plot inflow curve, often based on Darcy’s law with Vogel correction [Vogel] and pump intake curve. Pump intake
curve shows pressure difference in well from bottom to top including casing, pump and tubing.

Figure 2 Nodal analysis plot (from [Brown 1984])

In contrast to well system plot, nodal analysis plot focuses to well performance analysis instead of pump
performance. It depends on pump performance characteristics but uses indirectly and requires its conversion
from head to pressure drop. Calculating pressure drop from head developed by pump become complicated with
presence of free gas fraction at pump intake. Free gas decrease gas-liquid mixture density and decrease
pressure drop developed by pump even assuming ESP can produce its head. Following example from [Brown
1984], one can find that for some specific well increasing number of stages in ESP will lead to decreasing intake
pressure and increasing of rate (Figure 3).
4 SPE 171338

Inspection of Figure 3 shows that, beyond 5375 stbl/d, the number of stages as the horsepower
requirement increase very fast without significant gain into production rate [Brown 1984]. Same EPS’s behavior
can be found on nodal analysis plot Figure 2 Increasing number of stages by the same number give less gain into
solution production rate with increasing of overall stages number due to multiphase nature of flow in pump at
low intake pressure. However, well system plot (Figure 1) cannot provide this information because of using non-
modified pump characteristics. With same effectiveness, we could discus frequency change in ESP installation
instead of changing number of stages.

Figure 3 Possible production rate vs number of stages (from [Brown 1984])

We have used dependency on number stages here, following Brown’s book, to show that calculating
pump intake curves, taking into account multiphase flow is not quite a new concept. Moreover, Gabor has
provided detailed algorithm for pump intake curve calculation in his book [Gabor 2009 p. 276 -277], although he
has mentioned its complexity and necessity of computer software. However, for some reasons nodal analysis
SPE 171338 5

plot with ESP intake curve is not widely used in industry. For example, many of popular pump design and
analysis software can draw well system plot but cannot draw nodal analysis plot with ESP intake curve.

Below we will show an example of using pump intake curve and nodal analysis plot for analyzing well
working with high gas fraction at pump intake. Combined with ESP head degradation model in can predict ESP
behavior well.

It is well known that ESP behavior can highly affected by presence of free gas in pump. Short overview of
gas influence can be found is SPE 117414 for example. Field and laboratory studies show that ESP loses its
ability to develop pressure drop in presence of free gas [Drozdov 2008, Duran 2004].

Besides decrease in gas-liquid mixture density discussed above - ESP also lose its ability to develop head.
This phenomenon can be described as head degradation. In this papers ESP head degradation model presented
by [Bedrin 2008] is used due to its simplicity. It illustrated by Figure 5.

Figure 4 ESP head degradation from laboratory tests. Upper plot – head vs rate, middle plot – power vs.
rate, lower plot – efficiency vs. rate [Drozdov 2004]
6 SPE 171338

1,0
0,9
0,8
0,7
Dimensionless head

0,6
0,5
0,4
0,3
0,2
0,1
0,0
0% 20% 40% 60% 80% 100%
Gas fraction
Standard ESP Technology 1 Technology 2 Technology 3

Figure 5 ESP head degradation from field case study [Bedrin 2008]

Quadratic form of ESP head degradation model simplify ESP intake curve calculations. Following [Bedrin,
2008] degradation coefficient for standard ESP can be expressed like

K deg  9  вх  0.6  вх  1
2

K deg  вх
Where – dimensionless degradation coefficient, and – gas fraction at pump intake. Taking into
account ESP head degradation an algorithm proposed by [Gabor 2004] can be upgraded as presented at Figure
6. The only change here is applying correction for ESP head degradation.
SPE 171338 7

Ql = q1

Assume GLR a

Calculate ΔPtub

Calculate PIP

GLRa = GLR
Apply correction
and calculate
corrected PIP1

Calculate
Ksep and GLR

ABS(GLR - GLRa)

Calculate
ΔPcas

Calculate
FBHP

More rates?

Plot curve

Figure 6 Flowchart for calculating pump intake curve (adopted after [Gabor 2004])

The algorithm are as follow:

1. Assume some rate q1

2. Assume gas liquid ratio (GLR) valid for the flow in tubing

3. Calculate pressure gradient in tubing using some multiphase flow correlation (we used Ansari
correlation [Ansari 1994]). Pressure gradient calculation will give a discharge pressure.
8 SPE 171338

4. Calculate pump intake pressure from ESP characteristic. Stage-by-stage calculation taking into
account gas and oil volume factors required. It is assumed that ESP develop its head due to its
characteristic.

5. Apply head degradation correction to pump pressure drop according [Bedrin 2008]. It will give a
pump intake pressure.

6. Calculate gas separation efficiency at pump intake (natural and forced) by some algorithms.
Marquez correlation for separation efficiency [Marquez 2003] and gas separator test results
[Drozdov 2008] used. Using separation efficiency new GLR in tubing can be estimated

7. Compare difference between new and old GLR. If it is not small enough repeat calculation from
step 3. If it is small enough proceed next step.

8. Calculate bottom hole pressure using multiphase flow correlation.

9. Check if more rate points need to be calculated.

10. Plot the curve.

Nodal analysis plot for ESP well

300 without ESP head degradation
250

200

150

100

50

0
0.0 20.0 40.0 60.0 80.0 100.0 120.0
Liquid rate, m3/day

45 Hz pump intake VLP without ESP 50 Hz pump intake

55 Hz pump intake 60 Hz pump intake 65 Hz pump intake
IPR

Figure 7 Nodal analysis plot without ESP head degradation

SPE 171338 9

Nodal analysis plot for ESP well

300 with ESP head degradation
250

200

150

Назва ние оси

100

50

0
0.0 20.0 40.0 60.0 80.0 100.0 120.0
Liquid rate, m3/day

45 Hz pump intake VLP without ESP 50 Hz pump intake

55 Hz pump intake 60 Hz pump intake 65 Hz pump intake
IPR

Figure 8 Nodal analysis plot with ESP head degradation

Results of algorithm calculation are illustrated of Figure 7, Figure 8. Figure 8 show that ESP head
degradation cause that flowing bottom hole pressure for specific conditions cannot be decreased below
65 bar (approx. 1000 psi) by frequency change.

As it was mentioned at the beginning, the majority of Western Siberia well are working close to its
production potential, which in most cases limited by amount of free gas ESP can handle. For example –
Priobskoe field – one of the largest oil fields in Russia – average Pwf varies from 40 to 70 atm. which
corresponds to 20 - 40% free gas fraction at pump intake after natural and forced separation. In
addition, it is often observer for ESP equipped well that the changes in frequency do not affect Pwf and
rates according to affinity law as it can be predicted from well system curve built by ESP design software
(Figure 9). This behavior often confuse engineers and provoke using rule of thumbs instead of
engineering calculations. Method presented allows to describe this phenomena taking into account ESP
had degradation and correctly estimate target bottom hole pressure. Moreover -Figure 4 shows that ESP
efficiency decreases with increasing gas fraction at pump intake. This make reasonable to estimate
overall efficiency by ESP installation with different frequencies in order to improve energy efficiency of
the system.
10 SPE 171338

Роснефть - Пурнефтегаз
Южный Харампур / Kуст 12 / Скважина 891
№2. Замер
НКТ
Дебит нефти, воды и газа при стандартных условиях относительно времени
250 №1. Общий №3. Общий №4. Замер №5. Общий замер
№6.
Период №8. Период 40000
замер. замер НКТ Замер №7. Общий замер Замер дебита №9. Замер
НКТ
НКТ, затруб общего
закрыт. дебита

225 55 Гц 57 Гц 59 Гц 61 Гц

30000

200

нефти и воды [м3/сут]

m3/day
20000
175

Q, дебит газа [м3/сут]

150
10000

125

Q, дебит
0

Q,
100

75
-10000

50

-20000
25

0 -30000
0

0
:0

:0

:0

:0

:0
05

05

05

05

05
0

0
.1

.1

.1

.1

.1
20

21

22

23

24
Q, нефт ь [м3/сут ] Q, в ода [м3/сут ]
Время/time
Q, газ [м3/сут ] 50 per. Mov. Avg. (Q, нефт ь [м3/сут ])

50 per. Mov. Avg. (Q, в ода [м3/сут ]) 50 per. Mov. Avg. (Q, газ [м3/сут ])

57Гц
Hz 59Гц
Hz 61Гц
Hz
Роснефть - Пурнефтегаз Роснефть - Пурнефтегаз Роснефть - Пурнефтегаз
Южный Харампур / Kуст 12 / Скважина 891 Южный Харампур / Kуст 12 / Скважина 891 Южный Харампур / Kуст 12 / Скважина 891
Дебит нефти, воды и газа при стандартных условиях относительно времени Дебит нефти, воды и газа при стандартных условиях относительно времени Дебит нефти, воды и газа при стандартных условиях относительно времени
300 40000 300 40000 300 40000

275 275 275

30000 30000 30000

250 250
250

225 20000

Q, дебит нефти и воды [м3/сут]

225 20000
Q, дебит нефти и воды [м3/сут] 225 20000
Q, дебит нефти и воды [м3/сут]

200 200

Q, дебит газа [м3/сут]

200

Q, дебит газа [м3/сут]

10000
Q, дебит газа [м3/сут]

10000 10000
175
175
175

150 0
150 0
150 0

125
125
125 -10000
-10000
-10000 100
100
100
75 -20000
75 -20000
75 -20000
50
50
50 -30000
-30000
-30000 25
25
25
0 -40000
0 -40000

:02

:16

:31

:45

:00

:14

:28

:43

:57

:12
0 -40000

05

05

05

05

06

06

06

06

06

07
:02

:16

:31

:45

:00

:14

:28

:43

:57

:12

.10

.10

.10

.10

.10

.10

.10

.10

.10

.10
05

05

05

05

06

06

06

06

06

07
:02

:16

:31

:45

:00

:14

:28

:43

:57

:12

24

24

24

24

24

24

24

24

24

24
.10

.10

.10

.10

.10

.10

.10

.10

.10

.10
05

05

05

05

06

06

06

06

06

07

23

23

23

23

23

23

23

23

23

23
.10

.10

.10

.10

.10

.10

.10

.10

.10

.10

Время/time
22

22

22

22

22

22

22

22

22

22

Время/time
Время/time Q, нефть [м3/сут] Q, вода [м3/сут] Q, газ [м3/сут]
Q, нефть [м3/сут] Q, вода [м3/сут] Q, газ [м3/сут]
Q, нефть [м3/сут] Q, вода [м3/сут] Q, газ [м3/сут]

Figure 9 example of frequency change on well behavior (from [Bedrin 2008])

References

1. Ansari, A.M. et al.: SPE 28671 “A comprehensive Mechanistic Model for Upward Two-Phase
Flow in Wellbores” 1994

2. Bedrin V. G., Khasanov M. M., Khabibullin R. A., Krasnov V. A., Pashali A. A., Litvinenko K. V.,
Elichev V. A. , Prado M., High GLR ESP technologies comparison. Field test results, SPE 117414,
2008

3. Gabor T. Electrical submersible pump manual: design, operation and maintenance. Gulf
professional publishing, Oxford, UK, 2009

4. Duran, J; Prado, M.G.: “ESP Stages Air-Water Two-Phase Performance - Modeling and
Experimental Data”, paper SPE presented at the 2004 SPE ESP Workshop.

5. Drozdov A.N. The technology and technique of oil production by submersible pumps in the
complicated conditions: Teaching aid for universities.- Moscow: MAKS Press, 2008. (in russian)

6. Kermit E. Brown. The Technology of Artificial Lift Methods. – PennWellBooks. – 1984. – 448 p.
 SPE 171338 11

7. Marquez, R.A. and Prado, M.G., “A new correlation for predicting natural separation efficiency”.
Southwestern Petroleum Short Course. April 16-17, 2003. Lubbock Memorial Civic Center, Lubbock,
Texas

8. Vogel J.V. Inflow Performance Relationships for Solution-Gas Drive Wells. J. Pet. Tech. (January
1968), p. 83-93.

View publication stats