Fadairo2013 Gas Well Unloading

SPE 167552
An Improved Predictive Tool for Liquid Loading in a Gas Well

1
Fadairo Adesina SPE, 1Femi-Oyewole Damilola SPE and 2Falode Olugbenga, SPE
1 2
Covenant University Ota, Nigeria, University of Ibadan, Nigeria
Copyright 2013, Society of Petroleum Engineers
This paper was prepared for presentation at the Nigeria Annual International Conference and Exhibition held in Lagos, Nigeria, 30 July–1 August 2013.
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
As the search for natural gas becomes increasingly high due to its high demand worldwide, the oil and gas
industry is faced with the challenge of liquid loading in gas or condensate wells. It is imperative to properly
design and predict the operational parameters necessary for handling flow assurance challenges due to
simultaneous flow of gas with liquid. The model of Guo et al is the most recent systematic approach for
predicting liquid loading in gas well. However, it did not account for the accumulation and kinetic terms in the
momentum energy equation used to estimate bottom-hole pressure in a gas/oil/water/solid four phase flowing
well. The two neglected terms in Guo et al formulation have significant effects on the gas well operational
parameters such as the minimum gas flow rate for preventing liquid loading.
This paper presents an improved model that describes a systematic approach for estimating liquid loading in a gas
well without neglecting any term in the fundamental momentum equation. The results obtained showed that at the
early production time where initial transience at the onset of flow is experienced, the critical gas flow rate
obtained from the new model is lower than that predicted from Guo et al model due to inclusion of accumulation
term while at the later production time, the critical gas flow rate obtained becomes higher than that predicted from
Guo et al model and increases as the transient period elapses. Results further show that at some point during
production, the minimum energy required to lift liquids out of the wellbore is more than that required at the earlier
stage of production. The new model is reasonable, reliable and better when compare with Turner et al and Guo et
al models. It is useful for operators to refine their procedures and better manage the risk of liquid loading during
natural gas production.
Introduction
Liquid loading is refers to as the inability of a gas well to remove liquids which are produced along with gas from
a wellbore. This phenomenon is activated when the produced gas has lost its capacity to carry up the co-produced
fluids up the wellbore (Coleman et al, 1999). A typical gas well produces natural gas and most often it carries
either liquid or condensate in the form of mist. The associated liquids accumulate in the production well; hence a
2 SPE 167552
hydrostatic pressure is created, till the point where the gas well ceases flowing (Guo et al 2006). It is always
attributed to insufficiency of the velocity of the gas phase in the tubing in transporting the liquid up to the surface
and most often it impedes or even halts gas production. The liquid that accumulates at the bottom hole leads to the
formation of a hydrostatic column which in return forces a back pressure on the reservoir and can kill the well
ultimately (Guo et a, 2006, Veeken et al, 2003).
Liquid loading has been a challenge in meeting the ever increasing demand for natural gas which increases day by
day, as it reduces the ability of a gas well to produce a at a required rate over a particular period of time (Nosseir
2000) .The problem of liquid loading is one of the most profits hurting occurrence in matured gas production well
that causes large financial losses in oil and gas industry. A major point to note is that as the reservoir pressure
reduces, gas velocity also reduces and therefore liquid loading is the end result, which could ultimately kill the
well. Liquid load-up in gas wells is not always obvious; therefore, a thorough diagnostic analysis of well data
needs to be carried out in order to adequately predict the rate at which liquids will accumulate in the well.
Although this subject has been studied extensively but the results from previous investigators and the most
commonly applied model in the industry still has a high degree of inaccuracy, especially in predicting the
minimum gas flow rate required to unload liquids from the wellbore.
Turner et al. (1969) were the very first investigators in the prediction and analysis of the minimum gas flow rate
needed to prevent liquid loading. They proposed two models that can be used in the liquid removal analysis; the
film movement model and the entrainment drop model. The entrained drop model was developed based on the
velocity of liquid and also the maximum drop diameter with a critical Weber number 30.
According to Turner et al.(1969), Li M. (2001) gas will continuously remove liquids from the well until its
velocity drops to below the terminal slip velocity. The minimum gas flow rate (in MMcf/D) for a particular set of
condition (pressure and conduit geometry) was presented. They compared the results of the two models with the
field test data, and it was discovered that the entrained drop model gave a better result for the prediction of liquid
loading. They also posited that in order to insure removal of drops a 20% increase in gas velocity would be
enough to remove droplets continuously in order to prevent liquid loading. The main challenges with the turner
model were the problem of estimating the values of pressure and density of the fluid, and they did not make
provision for calculating gas pressure in a wellbore with multiphase flow.
Coleman et al (1991), modified the turner model using the Turner model but validating with field data of lower
reservoir and wellhead flowing pressures all below approximately 500 psia. They were convinced that a better
prediction could be achieved without a 20% upward adjustment to fit field data.
The most recent work carried out in the systematic prediction of liquid loading in gas wells was done by Guo et
al(2006). Their method provides production engineers with a systematic approach to predicting the minimum gas
production rate for the continuous removal of liquids from gas wells The outcome showed that Turner et al’s
(1969) method with 20%-adjustment still under-estimated the minimum gas velocity for liquid removal. Starting
from Turner et al.’s entrained drop model, their study has determined the minimum kinetic energy of gas that is
required to lift liquids. A 4-phase (gas, oil, water, and solid particles) mist-flow model was developed. Applying
SPE 167552 3
the minimum kinetic energy criterion to the 4-phase flow model resulted in a closed form analytical equation for
predicting the minimum gas flow rate.
The model of Guo et al (2006) is the most recent systematic approach for predicting liquid loading in gas well.
However, it did not account for the accumulation and kinetic terms in the momentum energy equation used to
estimate bottom-hole pressure in a gas/oil/water/solid four phase flowing well. The two neglected terms in Guo et
al (2006) formulation have significant effects on the gas well operational parameters such as the minimum gas
flow rate for preventing liquid loading.
This paper presents an improved model that describes a systematic approach for estimating liquid loading in a gas
well without neglecting any term in the fundamental momentum equation..
METHODOLOGY
This model seeks to improve on the Guo et al (2006) model using field data containing 106 test points used by
turner, The model simply involves the addition of the accumulation term and the kinetic energy to the governing
equation used by Guo et al (2006) in developing his four phase flow model.
The most commonly used models for liquid loading prediction in gas well is presented by Turner et al1
Turner et al.’s (1969) terminal slip velocity equation is expressed in U.S. field units as
/ /
.
/ / (1)
The corresponding critical rate can be expressed as
(2)
NEW MODEL DEVELOPMENT

Minimum Kinetic Energy: Kinetic energy per unit volume of gas can be expressed as7

(3)
This parameter has been used in the well-drilling industry todetermine the minimum required gas flow rate for
effectivelytransporting drill cuttings in boreholes. Substituting Eq. (1) into Eq.(3) gives an expression for the
minimum kinetic energy required to keep liquid droplets from falling7:
σ ρ ρ
0.026 (4)
If the value of drag coefficient Cd = 0.44 (recommended by Turner et al.1969) is used, and the effect of gas
density is neglected (a conservative assumption), Eq. (4) becomes:
0.04 σρ (5)
In gas wells producing water, typical values for water-gas interfacial tension and water density are 60 dynes/cm
and 65lbm/ft3, respectively. This yields the minimum kinetic energy value of 2.5lbf-ft/ft3. In gas wells producing
4 SPE 167552
condensate, typical values for condensate-gas interfacial tension and condensate density are 20 dynes/cm and
45lbm/ft3, respectively. This yields the minimum kinetic energy value of 1.2lbf-ft/ft3. The minimum gas velocity
required for transporting the liquid droplets upward is equal to the minimum gas velocity required for floating the
liquid droplets (keeping the droplets from falling) plus the transport velocity of the droplets, i.e.,
(6)
The transport velocity may be calculated on the basis of liquid production rate, geometry of the conduit, and
liquid volume fraction, which is difficult to quantify. Instead of trying to formulate an expression for the transport
velocity , this study uses as an empirical constant to lump the effects of non-stagnation velocity, drag
coefficients for solid spheres, and the critical Weber number established for drops falling in air. On the basis of
Turner et al.s (1969) work, the value of vtrhas been taken as 20% of vslin this study. Use of this value results in
1.2 (7)
Substituting Eqs (1) and (7) into Eq. (3) results in the expression for the minimum kinetic energy required for
transporting the liquid droplets as7:
0.0576 σρ (8)
For typical gas wells producing water, this equation yields the minimum kinetic energy value of 3.6lbf-ft/ft3. For
typical gas wells producing condensate, this equation gives the minimum kinetic energy value of 1.73lbf-ft/ft3.
These numbers imply that the required minimum gas production rate in water producing gas wells is
approximately twice of that in condensate-producing gas wells. In order to evaluate the gas kinetic energy Ek in
Eq. (3) at a given gas flow rate and compare it with the minimum required kinetic energy Ekmin Eq. (8), the values
of gas density ρ and gas velocity vg need to be determined. Expressions for ρ and vg can be obtained from ideal
gas law7:
ρ (9)
.
4.71 10 (10)
Substituting Eqs. (9) and (10) into Eq. (3) yields:
6.46 10 (11)
Equation (11) indicates that the gas kinetic energy decreases with increased pressure, which means that the
controlling conditions are bottom hole conditions where gas has higher pressure and lower kinetic energy. This
analysis is consistent with the observations from air-drilling operations where solid particles accumulate at bottom
hole rather than top hole. However, this analysis is in contradiction with Turner et al.’s (1969) results that
indicated that the wellhead conditions are, in most instances, controlling.
New Four-Phase Flow Model
In order to accurately predict the bottomhole pressure P in Eq. (11), a gas-oil-water-solid 4-phase mist-flow
model has been developed. Derivation of the model is shown in Appendix. According to the 4-phase flow model,
the flowing pressure P at depth L can be solved numerically from the following equation:
SPE 167552 5
(12)
Where
. . . .
cos Ɵ (13)
. . .
(14)
.
(15)
.
(16)
(17)
Ɵ
′ (18)
.
(19)
(20)
Comparison and Discussion

Field data containing 106 test points used by Turner et al (1969) were employed in this study to compare the new
method to both Turner et al’s method and Guo et al’s method. The wells produced gas and condensate and/or
water from either the tubing or annulus. Gas specific gravity of 0.6 was assumed for all the wells, Gas-condensate
and gas-water interfacial tensions were assumed to be 20 and 60dynes/cm respectively, wellhead temperature of
60F was used and the roughness of the conduit wall was assumed to be 0.000015in. All other required data are
givem by Turner et al (1969).
6 SPE 167552
14000 Turner et al model
12000
Tested Flow Rate (Mcf/d)
10000
8000
6000
4000
2000
0
0 2000 4000 6000 8000
Calculated Minimum Flowrate (Mcf/d)
Fig 1: Turner et al’s model – calculated minimum flow rae mapped against the test flow rate
12000
Guos model
10000
Test Flow Rate (Mscf/d)
8000
6000
4000
2000
0
0 2000 4000 6000 8000 10000 12000
Calculated Minimum Flowrate (Mscf/d)
Fig 1: Guo et al’s model – calculated minimum flow rae mapped against the test flow rate
SPE 167552 7
NEW MODEL
12000
10000
Test Flow Rate(Mcf/d)
8000
6000
4000
2000
0
0 2000 4000 6000 8000 10000 12000
Calculated Minimum Flow Rate(Mcf/d)
Fig 1: new model – calculated minimum flow rae mapped against the test flow rate
Figure 1 above shows Turner et al’s calculated minimum flow rates(after 20% adjustment) mapped against the
test flow rates.The diagonal line seperates the predicted unloaded and loaded regions with the upper being the
unloaded and the lower being the loaded regions.
Figure 2 shows the Guos model mapping the calculated minimum rates against the test flow rates.
Figure 3 simply shows the new model mapping the calculated minimum rates against the test flow rates.at time
100days.
Application
The new model can be used for predicting the minimum flow rate required for water and/or condensate removal in
various well conditions. The charts below simply represent the critical gas production rates for water and
condensate removal. For instance in figures 4, consider a gas velocity stream flowing through a conduit of 3.5
inches tubing diameter accompanied by a water flow rate of 0bbl/d and 40bbl/d respectively, it was observed that
at a higher water flow rate the minimum gas flow rate experiences a significant upsurge at different conduit
lengths in the life of the well as show in the figure. Also as the depth increases, the minimum gas flow rate
increases. The minimum gas flow rate obtained in fig. 4 is higher than that of fig.5. Hence, we deduce that as the
tubing diameter is increased the gas flow rate required to lift liquids will be higher than that required to lift when
using a smaller diameter. This is owing to the fact that an increase in diameter will lead to an increased surface
area, which inadvertently will require more energy for efficiency. The minimum gas flow rate can easily be
extracted from the plots at different conduit lengths in increasing progression throughout the life of the well.
8 SPE 167552
3.5 in Tubing,,0.70sg,Gas,300 psia wellhead pressure (100days)
5000
Critical Production Rate
4000
(Mscf/d)
3000
0 b/d
2000 10 b/d
20 b/d
1000 30 b/d
0
0 2000 4000 6000 8000 10000
Depth (ft)
Fig 4: Critical flow rate for removal in a 3.5in tubing at 300psia at 100days of production
2.375‐in,Tubing,0.70SG Gas,900‐psia Wellhead pressure
1200
Critical Production
1000
Rate(Mscf/d)
800 0 b/d
600 10 b/d
400 20 b/d
200
30 b/d
0
2000 4000 6000 8000 10000 40 b/d
Depth(ft)
Fig 4: Critical flow rate for removal in a 2.375in tubing at 900psia at 100days of production
Transient Curve
1300
1250
1200
1150
1100
Q (Mscf/d)
1050
1000
950 Guo's model
900 New Model
850
800
750
700
0 20 50 100 200 400 500 600 800 1000
Time (days)
Fig 6: minimum gas flow rate against production time

SPE 167552 9
Figure 6 attempts to capture the transient property or behavior experience in a gas velocity stream. It is general
assumed that the source of liquid in the flow stream is due to condensation of gases to liquids in the wellbore as
pressure and temperature decreases. Liquids fallback and accumulate downhole. As a result of the accumulation
of liquids downhole and rise in liquid column, the hydrostatic pressure increases and exerts pressure on the
formation. This accumulation majorly arises from the fact that at the onset of the gas production phase, it exhibits
a transient flow behavior, which may eventually stabilize at a particular time in the life of the well.
The figure depicts that the critical flow rate increases from 0 to 400days and then stablises above 400days of
production time. It is evident that there exist an initial transience at the onset of flow and further validates that at a
certain time during production; the transient period elapses and gives way to a more stable steady state flow. It
also shows that at some point during production, the minimum energy required to lift liquids out of the wellbore is
more than that required at the earlier stage of production. However, Guo’s Model simply assumes a constant flow
rate throughout production without taking the time factor into consideration.
10 SPE 167552
Tabl
Pressure Test load Turner Guos(calc) New mode Test Turners Guo New mdl Guos Turner New
725 775 779 885.16 1130.3145 near lu LOAD ED LOAD ED LOADED T T T
400 417 583 462.17 1603.42 near lu LOAD ED LOAD ED LOADED T T T
108 568 306 257.08 1379.82 near lu UNLOADED UNLOADED LOADED F F T
340 712 661 505.71 1321.90 near lu UNLOADED UNLOADED LOADED F F T
450 442 419 469.04 1321.51 near lu UNLOADED LOAD ED LOADED T F T
3607 1525 1156 1337.9 2096.54 Loaded UNLOADED UNLOADED LOADED F F T
3434 2926 1150 1296.450112 2551.8045 unloaded UNLOADED UNLOADED UNLOAD ED T T T
3773 2494 1158 1357.253247 2918.3303 ques UNLOADED UNLOADED LOADED ques ques ques
3340 2611 2412 2932.52 5045.0913 Loaded UNLOADED LOAD ED LOADED T F T
3472 2572 1085 1328.809334 5914.6145 unloaded UNLOADED UNLOADED LOADED T T F
8215 3472 1956 2976.479746 3242.7319 Loaded UNLOADED UNLOADED UNLOAD ED F F T
7950 4896 1941 2927.057073 3064.6864 ques UNLOADED UNLOADED UNLOAD ED ques ques ques
2182 5501 3767 4598.493303 4162.7252 Loaded UNLOADED UNLOADED UNLOAD ED F F F
1590 3009 3281 3418.464147 3445.342 Loaded LOAD ED LOAD ED LOADED T T T
1245 4441 4920 5784.857379 5351.1483 Loaded LOAD ED LOAD ED LOADED T F T
2352 6361 5641 6893.340144 6299.1638 Loaded UNLOADED LOAD ED UNLOAD ED T F F
2862 3024 5098 5647.075055 5065.8334 unloaded LOAD ED LOAD ED LOADED F F F
1102 1365 1419 1094.897491 2545.2909 Loaded LOAD ED UNLOADED LOADED F T T
552 1607 958 779.9430952 2414.483 near lu UNLOADED UNLOADED LOADED near lu near lu near lu
660 4300 6367 4757.973005 4286.3881 Loaded LOAD ED LOAD ED UNLOAD ED T T F
210 470 3248 2661.527839 0 Loaded LOAD ED LOAD ED UNLOAD ED T T
Table 1: Critical Rate Calculated from Different Models and Their Corresponding Loading Status Based on Turner et al (1969)
SPE 167552 11
Number of Incorrectly predicted wells
Turner 24
Guo 20
New mdl 13
True represents prediction of test equal model prediction
False represents prediction of test data not equal to model prediction
Table 2: Number of Incorrectly Predicted Wells by Different Model based on Turner et al Data
Critical Flow Rate Comparisons
14000
12000
Critical Flow Rtae(Mscf/d)
10000
8000
New Model
6000 test
turner
4000
guo
2000
0
3660
3540
3338
3455
3212
7405
2175
1520
1958
1993
2256
2003
1835
5056
1902
1895
2184
2783
2574
2611
2149
725
450
760
552
484
280
Pressure(Pwf)
Figure 7: Critical Rate from Different Models at Varying Pressure
Table 1 shows the application of the new model to the well data from Turner et al (1969) paper. The first two
columns in the table are from Turner et al1 paper while column three, column four and column five are estimated
values from Turner etal model (1969), Guo et al model (2006), and the new model respectively. The column six,
column seven column eight and column nine show the loading condition of the well obtained from the test value,
Turner et al, Guo et al and the new model respectively. The loading condition of the gas well is obtained by
comparing the critical rate from a model with the test rate. If the test rate is greater than critical rate from model, it
means the well is unloaded otherwise it is loaded up. The last three column of the table compare a model result to
a test status. If the prediction from the model is correct, T is marked and if the prediction is incorrect F is marked.
The numbers of incorrectly predicted wells are show in Table 2. It was reveals that the number of incorrectly
predicted well from Turner et al model is 24 while that of Guo et al is 20 incorrectly predicted wells. The new
model gives the lower number of incorrectly predicted wells that shows that is more reasonable and reliable than
the existing models. Figure 7 shows the compared results from different models in table 1 that is the test values,
calculated Turneret al results, Guo et al results and the new model results. The calculated results from the new
model are almost in line or closer to the test values than the existing models.
12 SPE 167552
CONCLUSION
The new model evident that there exist an initial transience at the onset of flow and further validates that at a
certain time during production; the transient period elapses and gives way to a more stable steady state flow. It
also shows that at some point during production, the minimum energy required to lift liquids out of the wellbore is
more than that required at the earlier stage of production. However, Guo’s et al model simply assumes a constant
flow rate throughout production without taking the production time factor into consideration.
Guo et al’s four phase flow model did not consider pressure drop due to accumulation and the kinetic energy term
hence the new model is more accurate and precise than the Guo et al’s model and hence gives a more real
estimate of the minimum flow rate required for liquid removal.
The validation of the new model with the test values, Turner et al model and Guo model revealed that the new
model is reasonable and reliable for predicting liquid loading.
RECOMMENDATIONS
In natural gas wells, the liquid loading menace has to be adequately detected and removal of same has to be done
promptly, in order to efficiently maximize production time. The various assumptions used in the modeling and
systematic prediction of gas well loading need to be reviewed.
Both the pressure drop due to friction and the pressure drop due to accumulation must be taken into consideration
to ensure accurate prediction and proper estimation of liquid load up in gas condensate wells especially. This will
attempt to help expunge to a higher degree the accumulated liquids or slugs found in the wellbore.
It is highly recommended that the New Model should be adopted for various industry applications to further
confirm its efficiency as well as viability.
Nomenclature
A = cross-sectional area of conduit,ft2
Ai =Cross sectional area of conduit in2
Cd = drag coefficient
Dh =Hydraulic diameter,in
DH =Hydraulic diameter,ft
Ek=Gas specific Kinetic energy,ibf-ft/ft3
f = moody friction factor
g = gravitational acceleration,32.17ft/s2
gc =Unit conversion factor,ibm-ft/ibf-s2
L = Conduit length
P =pressure psia
Phf = wellhead flowing pressure psia
P =pressure ibf/ft2
qg = gas flow rate ft3/s
qo = oil flow rate ft3/s
qs = solid flow rate ft3/s
qw = water flow rate ft3/s
Qg=gas production rate ,Mscf/d
Qo =Oil production rate Mscf/d
Qs = solid production rate Mscf/d
SPE 167552 13
Sg =specific gravity of gas,air =1

So =specific gravity of produced oil,fresh water =1
Ss = Specific gravity of produced solid,fresh water =1
Sw = Specific gravity of produced water,fresh water =1
T = temperature R
Vg = Gas velocity ft/s
Vgm = Minimum gas velocity required to transport liquid drops,ft/s
REFERENCES
1. Coleman, S.B., Clay, H.B., McCurdy, D.G. and Norris III, H.L., “A New Look at Predicting Gas‐Well Load‐
Up,” paper SPE 20280, JPT (March 1991)
2. Fadairo A.S, Ako C, Falode O., “Modeling Productivity Index for Long Horizontal Well” Journal of Energy
Resource Technology, Vol 133, Issue 3, 2011.
3. Guo, Boyun and Ali Ghalambor. “A systematic approach to predicting liquid loading in gas wells. SPE 94081,
February 2006.
4. Li, M., Li, S.L. and Sun, L.T., “New View on Continuous‐Removal Liquids From Gas Wells,” paper SPE
75455, presented at the 2001 Permian Basin Oil and GasRecovery Conference, Midland, Texas, May 15‐16
5. Nosseir, M.A., Darwich, T.A., Sayyouh, M.H. and El Sallaly, M., “A New Approach forAccurate Prediction of
Loading in Gas Wells Under Different Flowing Conditions,”paper SPE 66540, SPE Production & Facilities Vol.
15 Number 4 (November 2000)
6. Turner, R.G., Hubbard, M.G. and Dukler, A.E., “Analysis and Prediction of Minimum Flow Rate for
Continuous Removal of Liquids from Gas Wells,” paper SPE 2198, JPT (November 1969)
7. Veeken, K., Bakker, E., and Verbeek, P., “Evaluating Liquid Loading Field Data andRemedial Measures,”
paper presented at the 2003 Gas Well De‐Watering Forum, Denver, CO, March 3‐4
APPENDIX
MODEL DERIVATION
For the development of the four-phase flow model;
GOVERNING EQUATION
Consider a mist-flow of gas, oil, water, and solid particles flowing upward in a conduit. The pressure
incrementdPover a small section of conduit dLwith an inclinationangle of θ consists of four components:
14 SPE 167552
hydrostatic pressurechange owing to the weight of the fluid, the pressure drop owing to friction, the pressure drop
owing to accumulation and kinetic energy term (Fadairo et al 2011). The following relation can be formulated on
thebasis of the first law of thermodynamics:
cos (A-1)
where
P = pressure, lb/ft2
= specific weight of mixture, lb/ft3
θ = inclination angle, degree
f = Moody friction factor, dimensionless
vmx = mixture velocity, ft/sec
gc = 32.2 ft/sec2
Dh = hydraulic diameter of the conduit, ft
L = conduit length, ft
The specific weight of the mixture is expressed as:
Ẇ Ẇ Ẇ Ẇ
(A2)
where
Ẇ = solid particle weight flow rate, lb/sec
Ẇ = water weight flow rate, lb/sec
Ẇ = oil weight flow rate, lb/sec
Ẇ = gas weight flow rate,lb/sec
qs = solid particle volumetric flow rate,ft3/sec
qw = water volumetric flow rate,ft3/sec
qo = oil volumetric flow rate, ft3/sec
qg = in-situ gas volumetric flow rate, ft3/sec
The weight flow rate of solid particles can be expressed in termsof daily sand/coal particle production rate:
.
Ẇ (A-3)
Where:
Qs = sand production rate in ft3/D
Ss= solid specific gravity, fresh water = 1.
The weight flow rate of water can be expressed in terms of daily water production rate:
.
Ẇ 62.4 (A-4)
Where:
Sw = specific gravity of produced water, fresh water = 1
Qw = water production rate, B/D.
SPE 167552 15
The weight flow rate of oil (condensate) can be expressed in terms of daily oil production rate:
.
Ẇ 62.4 , ( A-5)
Where:
So = specific gravity of produced oil, fresh water = 1
Qo = oil production rate, STB/D.
The weight flow rate of gas depends on gas production rate, Qg , and the specific gravity of gas:
Ẇ 0.0765 , (A-6)
Where:
Sg = gas specific gravity, air = 1
Qg = gas production rate, Scf/D.
The volumetric flow rates of solid, water, oil, and gas are expressed in field units as:
(A-7)
.
(A-8)
.
(A-9)
.
(A-10)
respectively, where T is the in-situ temperature in °R.

Substituting, Eqs. A-3 through A-10 into Eq. A-2 gives
. . .
. . .
. . . (A-
11)
which can be simplified as:
(A-12)
Where
. . . .
(A-13)
and
. . .
(A-14)
The mixture velocity can be formulated based on the total volumetricflow rate given by Eqs. A-7 through A-10
and conduitcross-sectional area:
. . .
(A-15)
Where, A is conduit cross-sectional area in ft2. Eq. A-15 can bewritten as

(A-16)
Where
16 SPE 167552
.
(A-17)
And
.
. (A-18)
(A-19)
Substituting Eqs A-12 and A-19 into Eq. A-1 yields the governingequation:

cos (A-20)
e (A-21)
cos (A-22)
and
(A23)
Ɵ
(A24)
.
(A25)
(A26)
SOLUTION
Using the average temperature over the conduit length, separation of variables on Eq. A-20 results in
(A-27)
Applying the boundary condition P=Phf(wellhead flowing pressure)at L=0 (wellhead) and integrating over
conduit length yields
(A-28)
Eq. A-28 can be integrated as:

^ ^ ^ ^ ^
[ ] + {( ^2 ∗
^ ^
)}
^ ^
[(tanh tanh *
2 b+2 2 2 4
3 4 3 4 4 8 4
2 2 ) (A29)

Fadairo2013 Gas Well Unloading

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fadairo2013 Gas Well Unloading

Uploaded by

Copyright:

Available Formats

SPE 167552

An Improved Predictive Tool for Liquid Loading in a Gas Well

Copyright 2013, Society of Petroleum Engineers

The corresponding critical rate can be expressed as

NEW MODEL DEVELOPMENT

Substituting Eqs. (9) and (10) into Eq. (3) yields:

Comparison and Discussion

Fig 6: minimum gas flow rate against production time

Figure 7: Critical Rate from Different Models at Varying Pressure

Sg =specific gravity of gas,air =1

respectively, where T is the in-situ temperature in °R.

Where, A is conduit cross-sectional area in ft2. Eq. A-15 can bewritten as

Eq. A-28 can be integrated as:

You might also like