PETROLEUM EXPLORATION AND DEVELOPMENT

Volume 45, Issue 2, April 2018

Online English edition of the Chinese language journal

Cite this article as: PETROL. EXPLOR. DEVELOP., 2018, 45(2): 351–357. RESEARCH PAPER

Prediction of wax precipitation region in wellbore during

deep water oil well testing
GAO Yonghai1, 2, LIU Kai1, ZHAO Xinxin1, 2, LI Hao1, 2, CUI Yanchun1, XIN Guizhen1, SUN Baojiang1, 2, *
1. School of Petroleum Engineering, China University of Petroleum, Qingdao 266580, China;
2. National Engineering Laboratory for Testing and Detection Technology of Subsea Equipments, Qingdao 266580, China

Abstract: During deep water oil well testing, the low temperature environment is easy to cause wax precipitation, which affects the normal
operation of the test and increases operating costs and risks. Therefore, a numerical method for predicting the wax precipitation region in
oil strings was proposed based on the temperature and pressure fields of deep water test string and the wax precipitation calculation mod-
el. And the factors affecting the wax precipitation region were analyzed. The results show that: the wax precipitation region decreases
with the increase of production rate, and increases with the decrease of geothermal gradient, increase of water depth and drop of water-cut
of produced fluid, and increases slightly with the increase of formation pressure. Due to the effect of temperature and pressure fields, wax
precipitation region is large in test strings at the beginning of well production. Wax precipitation region gradually increases with the increase
of shut-in time. These conclusions can guide wax prevention during the testing of deep water oil well, to ensure the success of the test.

Key words: deep water oil and gas development; oil well testing; wellbore wax precipitation; temperature field; pressure field; wax
precipitation region prediction

Introduction tion temperature at a given pressure or the wax precipitation

Well testing is an important part in the oil and gas explora-
tion which can provide reliable data for the evaluation of oil
and gas reservoirs and the selection of development mode[12].
However, in the test operation of deep-water oil wells, the
special low temperature environment is likely to cause wax
precipitation in the wellbore. When the temperature of the
crude oil drops to the initial crystallization temperature of the
wax during well testing, wax crystal particles begin to pre-
cipitate[3]. The precipitation of wax will increase the fluid
viscosity in the wellbore and increase the pressure loss[4]. The
wax crystal precipitation and deposition on the wall of the
wellbore will reduce the effective diameter of tubing and in-
crease the frictional resistance[3,56], and in turn influence the
accuracy of the test data and the efficiency of test operation.
Researchers in China and abroad have done a lot of studies
on the prediction of wax precipitation region and the calcula-
tion of wax deposition rate in waxy oil pipeline and wellbore
during production. The wax thermodynamic model plays an
important role in predicting the wax precipitation region.
Combining with the crude oil components, the wax thermo-
dynamic model can be used to determine the wax precipita-
pressure at a given temperature[7]. A lot of experimental and
theoretical studies have been carried out to based on the wax
thermodynamic model and a series of mathematical models
have been proposed to calculate the conditions of wax pre-
cipitation. The wax thermodynamic models can be roughly
divided into four types: regular solution model[811], polymer
solution model[1215], state equation model[1618] and crude oil
composition model[1921]. Nowadays, the understanding of
wax deposition rate in single-oil phase flow has been mature.
But the study on wax deposition rate under the gas-liquid
two-phase flow is still at the initial stage, and mainly focuses
on the case of horizontal pipeline[2226]. Moreover, there is less
research on wax deposition rule in the gas-liquid two-phase
wellbore flow.
In this study, a numerical method was proposed to predict
the wax precipitation region during deepwater oil well testing
based on the temperature and pressure fields calculation mod-
el combining the conditions of wax precipitation. And the
effects of production rate, geothermal gradient, water depth,
formation pressure, water-cut of produced fluid, and well
shut-in and opening etc. on the wax precipitation region are
analyzed.

Received date: 25 Sep. 2017; Revised date: 11 Jan. 2018.

* Corresponding author. E-mail: sunbj1128@126.com
Foundation item: Supported by the National Key Basic Research and Development Program (973 Program), China (2015CB251205).
Copyright © 2018, Research Institute of Petroleum Exploration & Development, PetroChina. Publishing Services provided by Elsevier B.V. on behalf of KeAi Com-
munications Co., Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
 GAO Yonghai et al. / Petroleum Exploration and Development, 2018, 45(2): 351–357

1. Calculation models for temperature and (2)

pressure fields and wax precipitation  1  
  l l     Al lvl   m l  e   d  Gl
t A s l  d
1.1. Temperature and pressure calculation models
(3)

The distributions of temperature and pressure in the test
t
 g gvg + d lvd     g +d  ps 
string are important factors affecting the wax precipitation
1  1 S
region. By solving the basic model of gas-liquid two-phase
flow, the distribution of temperature and pressure in the test A s
 A g  g vg2  A d  lvd2   g g vg vg g 
2 4A
1 Sin
+   g  g   d l  g cos  +
string can be obtained. Four types of the basic models of gas-
in g vr vr
liquid two-phase flow are: homogeneous flow model, sepa- 2 4A
rated flow model, drift-flux model and double-fluid model[2728]. l
m va   ev in  d v d (4)
In the homogeneous flow model, the gas-liquid two-phase l  d
mixed fluid is assumed as homogeneous medium, and its  p 1 
characteristic parameters are calculated according to the mean t
  l lv l     l 
s A s
 Al lvl2   12 l lvl vl 4SAl 
value of the gas-liquid two-phase medium. In the separated 1 S l
in g vr vr in   l l g cos    m v a  ev in 
flow model, the differences between different phases are taken 2 4A l  d
into account, the gas and liquid phases are treated separately,  l
and the fluids of different phases are considered to have inde-  d vd   l D   l  g  g sin  (5)
s
pendent flow rate and physical parameters. The drift-flux   1   1 
 g  g  Eg  vg2  gh    l l  El  vl2  gh  
model considers not only the relative motion between the gas t   2   2 
and the liquid phases, but also the distribution of the void
 1     1 
fraction and velocity along the cross section. The double-fluid  d  d  Ed  vd2  gh      g g vg  H g  vg2  gh  
 2  s   2 
model regards each phase fluid as continuous medium, and
establishes continuity equations, momentum conservation  1   1 
 l lvl  H l  vl2  gh    d d vd  H d  vd2  gh   H s  U
equations and energy conservation equations for the gas and  2   2 
liquid phases respectively, which are coupled with each other (6)
through the action of phase interface. In these types of models, In Eqs.1-6, there are seven basic unknown variables, p, T,
the homogeneous flow model is simple in calculation and vg, vl, βg, βl and βd, and other unknown variables can be ex-
convenient in application, but without considering the differ- pressed or solved by these basic variables. The pressure equa-
ence between gas phase and liquid phase, leading to large tion can be expressed as:
error. The flow characteristics of each phase in the separated   g  g 1   g l  p 1 
flow model are isolated. The drift-flux model cannot accu- 
 p

  p
  
t Ag s
 Ag gvg  
 g l 
rately represent the movement and spatial distribution of eve-
1  1 
ry phase. The double-fluid model is always appropriate as  Al lvl    Ad lv d  
Al s A l s
long as the assumption that the gas-liquid phase is continuous
medium in the local area is met. It is suitable for any binary  1 1 1 1 1
m 
  Gg  Gl  Gd
  g l 
(7)
mixture that can be considered as continuous medium and can   g l l
be used for a variety of flow patterns. The established equa- Eqs.1-7 can be simplified as functions of the seven basic
tion including comprehensive content, rich parameters and unknown variables and written in the form of Eq. 8.
extensive application[28]. p T v v   
In this study, the modified double-fluid model was used to a1  a2  a3 g  a4 l  a5 g  a6 l  a7 d 
t t t t t t t
calculate the temperature and pressure fields. By solving three p T v g vl  g  l  d
separate mass conservation equations (respectively on gas b1 +b2 +b3 +b4  b5 +b6 +b7 =c
s s s s s s s
phase, liquid droplet and wall film, as Eqs.1-3), two momen- (8)
tum equations (respectively on gas phase and liquid droplets
mixture, liquid film, as Eqs. 4-5) and one mixture energy 1.2. Calculation model of wax precipitation
conservation equation (as Eq. 6), the parameters of the multi- The wax precipitation conditions were calculated by the
phase flow can be calculated. The above equations were ob- extended model proposed by Rønningsen et al.[30] based on the
tained according to the law of conservation on the basis of the Pedersen model[16].
double-fluid model, and selecting the microelements[29]. When the liquid phase (oil) and the solid phase (wax) reach
 1 
t
  g g   
A s
 Ag gvg   m  Gg (1) the thermodynamic equilibrium state, the fugacity of a com-
ponent in the liquid phase and the solid phase is equal, as
 1   shown in Eq. 9:
  d l     A d lv d   m d   e  d  Gd
t A s l  d f li  fs (9)
 352 
 GAO Yonghai et al. / Petroleum Exploration and Development, 2018, 45(2): 351–357

The fugacity of the liquid phase and solid phase can be ex- 2.4. Method for wax precipitation region prediction
pressed as follows, respectively:
Combined with calculation models of temperature, pressure
f li  xlili p (10)
and wax precipitation, wax precipitation region in test string
 T  V  p  pref  
 H f can be predicted. The prediction method is as follows: (1) By
f s  xsoli p exp  
1     (11)
 RT
 Tf  RT  solving the temperature and pressure calculation models, the
distribution curves of fluid temperature (Fig. 1a) and the
where ΔHf and Tf in Eq.11 were calculated by Won’s for-
pressure along the well depth (Fig. 1b) in the test string are
mula[8].
obtained. (2) The temperature and pressure curve of wax pre-
2. Model solution and prediction method of wax cipitation (Fig. 1c) is obtained from the model of wax pre-
precipitation region cipitation. Combined with the well depth-pressure curve, tak-
2.1. Initial conditions ing the pressure as the intermediate variable, the wax precipi-
tation temperature-well depth curve (Fig. 1d) is obtained. (3)
During the flow test, the initial conditions are the tempera- Comparing the fluid temperature-well depth curve of the test
ture and pressure at the beginning of the test. The temperature string and the wax precipitation temperature-well depth curve,
in the test string is the ambient temperature, and the sum of the area enclosed by the two curves is the wax precipitation
the back pressure at the wellhead and the static pressure of region (Fig. 1e).
test fluid is the initial pressure condition.
3. Factors affecting wax precipitation region in
T j 0  Taj
 (12) the wellbore during deep-water oil well testing
 p j 0  po   t gh j
Based on the temperature and pressure calculation models
At the stage of well shut-in, the initial temperature in the
and wax precipitation condition calculation model, combined
test string is the temperature when the flow is stable. The ini-
with data from Well Louro-2 in Angola, the factors affecting
tial pressure condition is the sum of the wellhead back pres-
the wax precipitation region in deep water oil well testing
sure and the static pressure of the produced fluid when the test
were analysed. Well Louro-2 is a vertical deep-water oil well.
is stable.
The basic data includes: the design depth of 3 658.3 m, the
T j 0  Tsj water depth of 1 892 m, the sea surface temperature of 26 C,
 (13)
 p j 0  pso   o gh j the seabed temperature of 4 C, the geothermal gradient of
2.92 C/100 m, and the formation pressure and temperature of
2.2. Boundary conditions 40 MPa and 55.576 C, respectively, the production rate of
The boundary conditions of the flow test are as follows: at a 200 m3/d. The reservoir fluid components are shown in Table
certain rate of production, the bottom hole flowing pressure is 1, and the wellbore structure is shown in Fig. 2. When ana-
constant, and the bottom hole fluid temperature is equal to the lyzing a certain influence factor, taking different values of the
formation temperature. factor, and using basic data for the other parameters.
The boundary conditions during well shut-in are as follows:
the temperature of bottom hole fluid is equal to the tempera-
ture of formation, and the bottom hole pressure is the forma-
tion pressure.

2.3. Model solution method

The models for the temperature and pressure calculation in

the test string are complex nonlinear equations, which needs
to be solved by the numerical calculation method in combina-
tion with the auxiliary equations of initial conditions and
boundary conditions. Eqs. 1-7 can be simplified into the form
of Eq. 8, and then form a set of equations. It can be solved
based on the calculation of each coefficient, and references
[2829] present the solution procedure in detail. According to
the finite volume method, staggered spatial meshes were ap-
plied to discretize the testing string into several elements. The
velocity, mass flow and other flow variables were stored on
the boundary of the elements, and the pressure, temperature,
mass and other volume variables were stored in the center of
the elements. Then, the Euler method was used for numerical Fig. 1. Schematic diagram of method for wax precipitation re-
calculation by using the donor cell difference scheme. gion prediction in deep water oil well testing.
 353 
 GAO Yonghai et al. / Petroleum Exploration and Development, 2018, 45(2): 351–357

Table 1. Composition of reservoir fluid in Well Louro-2, Angola. because that the flow velocity of the upward fluid in the test
Component Content/% Component Content/% string is higher under the higher production rate, so there is
less heat exchange between the produced fluid and the sur-
N2 0.001 nC5 1.637
rounding environment and thus smaller temperature drop. In
CO2 0.482 C6 3.707
addition, the change of the wax precipitation temperature
CH4 34.978 C7 3.800 curve caused by pressure change is not significant in the range
C2H6 2.813 C8 3.477 of production. Therefore, the region where the output fluid
C3H8 4.008 C9 3.530 temperature is higher than the wax precipitation temperature
iC4 1.026 C10 3.480 is large. When the production rate is low, the heat exchange
nC4 2.774 C11+ 32.510 between the fluid in the test string and the low ambient tem-
iC5 1.777 perature is sufficient, which results in large drop in fluid tem-
perature and increase of the wax precipitation risk.
3.2. Geothermal gradient
As can be seen from Fig. 4, the wax precipitation region
decreases with the increase of geothermal gradient. The rea-
son is that the temperature of the tested zone is high when the
geothermal gradient is high, so the temperature of fluid flow-
ing into the wellbore from the reservoir is high. In addition, as
the temperature of the formation is also high, the heat loss of
the output fluid is small during upward flow from the bottom
of the well, leading to smaller temperature drop, and the tem-
perature in the whole test string is high, so the wax precipita-
tion region is smaller.
3.3. Water depth
It can be seen from Fig. 5 that the wax precipitation region
increases with the increase of water depth. The reason is that
when the water depth increases, the temperature of the area
near the seabed mud line becomes lower, moreover, the low
temperature section of the sea water increases. Therefore,
Fig. 2. Schematic diagram of well structure of Well Louro-2, more heat is dissipated to the outside in the process of upward
Angola. flow of the oil, which causes fast and large drop of the pro-
duced fluid temperature, and subsequently wax precipitation
3.1. Production rate
and expansion of the wax precipitation region.
As shown in Fig. 3, it is obvious that the wax precipitation
3.4. Formation pressure
region decreases with the increase of production. Wax pre-
cipitation will not occur in the whole test string when the It can be seen from Fig. 6 that the wax precipitation region
production rate increases to more than 300 m3/d. This is slightly increases with the increase of formation pressure. This

Fig. 3. Prediction of wax precipitation region at various pro- Fig. 4. Prediction of wax precipitation region at various geo-
duction rates. thermal gradients.

 354 
 GAO Yonghai et al. / Petroleum Exploration and Development, 2018, 45(2): 351–357

Fig. 5. Prediction of wax precipitation region at various water

depths. Fig. 7. Prediction of wax precipitation region at various wa-
ter-cuts.

string is equal to the static pressure of the test fluid, and the
temperature in the test string is lower, approaching the ambi-
ent temperature (Fig. 8). At this stage, the wax may precipitate
in a large region when oil starts to flow (Fig. 8).
There is still high pressure in the test string after well
shut-in. It can be seen from Fig. 9 that in the early stage of

Fig. 6. Prediction of wax precipitation region at various forma-

tion pressures.

is because that the pressure in the test string increases with the
increase of the formation pressure. According to the wax pre-
cipitation model, wax precipitation temperature increases with
the increase of pressure when the pressure is higher than the
fluid saturation pressure. Meantime, the change of formation
Fig. 8. Prediction of wax precipitation region at the beginning
pressure has insignificant influence on the temperature distri- of well opening.
bution in the test string, so the wax precipitation region in-
creases slightly .
3.5. Water-cut
As shown in Fig. 7, the water-cut of output fluid has insig-
nificant effect on the temperature distribution of the test string
in the formation, however it has significant effect on the fluid
temperature of the test string in the sea water. With the in-
crease of water-cut of produced fluid, the wax precipitation
region decreases. Because the specific heat capacity of water
is higher than that of the crude oil, and the high water-cut
reduces the temperature drop during the upward flow of the
fluid, resulting in the higher fluid temperature in the whole
test string and smaller wax precipitation region.
3.6. Well opening and shut-in
Fig. 9. Prediction of wax precipitation region at different
At the beginning of well opening, the pressure in the test shut-in time.
 355 
 GAO Yonghai et al. / Petroleum Exploration and Development, 2018, 45(2): 351–357

well shut-in, the temperature of the fluid in the test string de- T—temperature, K;
creases rapidly due to the large difference of temperature be- Taj—formation or sea water temperature at microelement j, K;
tween fluid in test string and the external environment. With Tf—melting temperature, K;
the increase of shut-in time, the temperature drop gradually Tj0—initial temperature of microelement j, K;
decreases due to the reduction of temperature difference be- Tsj—temperature in test string of the microelement j when test is
tween them, and the fluid temperature in the test string gradu- stable, K;
ally decreases and is close to the ambient temperature. Wax U—heat transfer per unit volume per unit time, J/(m3·s);
precipitation region gradually increases with the increase of v—velocity, m/s;
shut-in time. Therefore, this stage should be given close atten- vr—relative velocity, m/s;
tion and corresponding measures should be taken to reduce va—velocity of the part within phase transition, m/s;
the risk of wax precipitation. xli, xs—mole fraction of liquid phase, solid phase, %;
α—angle between the pipe and the vertical direction, rad;
4. Conclusions
β—volume fraction, %;
During the deep-water oil well testing, wax precipitation ΔV—difference of molar volume between the solid and liquid
region decreases with the increase of production rate in the phase, m3/mol;
flow test stage, and the lower production rate will increase the φli—liquid phase fugacity coefficient;
risk of wax precipitation. Lower geothermal gradient, deeper φoli—liquid phase fugacity coefficient at standard state;
water depth and lower water-cut will significantly increase the λ—friction coefficient;
wax precipitation region. The formation pressure has insig- ρ—density, kg/m3;
nificant effect on the wax precipitation region, and the wax ρo—density of produced fluid, kg/m3;
precipitation region increases slightly with the increase of the ρt—density of test fluid, kg/m3;
formation pressure. A larger wax precipitation region is ob- ψm—mass transfer rate between the phases, kg/(m3·s);
served at the beginning of well opening due to the effect of ψe—entrainment rate of liquid droplet, kg/(m3·s);
low temperature and high pressure fields. With the increase of ψd—deposition rate of liquid droplet, kg/(m3·s).
shut-in time, the fluid temperature in the test string decreases
gradually until close to the ambient temperature, and the wax Subscripts:
precipitation region also increases obviously with the decrease d—droplet;
of temperature. g—gas;
in—gas-liquid interface;
Nomenclature l—liquid film at wall surface.

a1, a2, a3, a4, a5, a6, a7, b1, b2, b3, b4, b5, b6, b7, c—coefficient; References
A—pipe cross-sectional area, m2;
D—pipe diameter, m; [1] MEN Xiangyong, YAN Xia, CHEN Yongchang, et al.
E—internal energy per unit mass, J/kg; Gas-water phase flow production stratified logging technol-
fli, fs—fugacity in the liquid phase, solid phase respectively, Pa; ogy of coalbed methane wells. Petroleum Exploration and
g—gravity acceleration, 9.8 m/s2; Development, 2017, 44(2): 289–294.
G—possible mass source, assumed to enter into the pipe at [2] MOGBO O. Deepwater DST design, planning and opera-
90°angle with the pipe wall, kg/(m3·s); tions: Offshore Niger Delta experience. SPE 133772, 2010.
h—well depth, m; [3] BANKI R, HOTEIT H, FIROOZABADI A. Mathematical
hj—vertical depth of microelement j, m; formulation and numerical modeling of wax deposition in
H—enthalpy per unit mass, J/kg; pipelines from enthalpy- porosity approach and irreversible
Hs—enthalpy from mass sources per unit volume per unit time, thermodynamics. International Journal of Heat & Mass Trans-
J/(m3·s); fer, 2008, 51(13): 3387–3398.
ΔHf—melting enthalpy at the normal melting point, J/mol; [4] RIBEIRO F S, MENDES P R S, BRAGA S L. Obstruction
p—pressure, Pa; of pipelines due to paraffin deposition during the flow of
pj0—initial pressure of microelement j, Pa; crude oils. International Journal of Heat & Mass Transfer,
po—wellhead back pressure, Pa; 1997, 40(18): 4319–4328.
pref—pressure under reference state, Pa; [5] KELECHUKWU E M, AL-SALIM H S, SAADI A. Prediction
pso—wellhead back pressure when test is stable, Pa; of wax deposition problems of hydrocarbon production sys-
R—gas constant, 8.314 J/(mol·K); tem. Journal of Petroleum Science & Engineering, 2013, 108:
s—length, m; 128–136.
S—wetted perimeter, m; [6] DUAN J, WANG W, DENG D, et al. Predicting temperature
t—time, s; distribution in the waxy oil-gas pipe flow. Journal of Petro-

 356 
 GAO Yonghai et al. / Petroleum Exploration and Development, 2018, 45(2): 351–357
leum Science & Engineering, 2013, 101: 28–34.
[7] MEI Haiyan, KONG Xiangyan, ZHANG Maolin, et al. A
thermodynamic model for the prediction of paraffin deposi-
tion. Petroleum Exploration and Development, 2000, 27(1):
84–86.
[8] WON K W. Continuous thermodynamics for solid-liquid equi-
libria: Wax formation from heavy hydrocarbon mixtures//
Proceedings of AIChE Spring National Meeting. Houston,
Texas: American Institute of Chemical Engineers, 1985:
24–28.
[9] PEDERSEN K S, SKOVBORG P, RONNINGSEN H P. Wax
precipitation from North Sea crude oils. 4. Thermodynamic
modeling. Energy and Fuels, 1991, 5(6): 924–932.
[10] ERICKSON D D, NIESEN V G, BROWN T S. Thermody-
namic measurement and prediction of paraffin precipitation in
crude oil. SPE 26604, 1993.
[11] THOMAS F B, BENNION D B, HUNTER B E. Experimental
and theoretical studies of solids precipitation from reservoir fluid.
Journal of Canadian Petroleum Technology, 1992, 31(1): 22–31.
[12] HANSEN J H, FREDENSLUND A, PEDERSEN K S, et al. A
thermodynamic model for predicting wax formation in crude
oils. AIChE Journal, 1988, 34(12): 1937–1942.
[13] ZHOU X, THOMAS F B, MOORE R G. Modelling of solid
precipitation from reservoir fluid. Journal of Canadian Petro-
leum Technology, 1996, 35(10): 37–45.
[14] PAULY J, DAUPHIN C, DARIDON J L. Liquid-solid equi-
libria in a decane+multi-paraffins system. Fluid Phase Equi-
libria, 1998, 149(1/2): 191–207.
[15] COUTINHO J A P, RUFFIER-MÉRAY V. Experimental
measurements and thermodynamic modeling of paraffinic
wax formation in undercooled solutions. Industrial & Engi-
neering Chemistry Research, 1997, 36(11): 4977–4983.
[16] PEDERSEN K S. Prediction of cloud point temperatures and
amount of wax precipitation. SPE Production & Facilities,
1995, 10(10): 46–49.
[17] HANSEN A B, LARSEN E, PEDERSEN W B, et al. Wax
precipitation from North Sea crude oils. 3. Precipitation and
dissolution of wax studied by differential scanning calo-
rimetry. Energy & Fuels, 2012, 5(6): 914–923.
[18] LIRA-GALEANA C, FIROOZABADI A, PRAUSNITZ J M.
 357 
Thermodynamics of wax precipitation in petroleum mixture.
AIChE Journal, 1996, 42(1): 239–248.
[19] PAN H, FIROOZABADI A, FOTLAND P. Pressure and
composition effect on wax precipitation: Experimental data
and model results. SPE Production & Facilities, 1997, 12(4):
250–258.
[20] LEONTARITIS K J. PARA-based (Paraffin-Aromatic-Resin-
Asphaltene) reservoir oil characterizations. SPE 37252-MS,
1997.
[21] LEELAVANICHKUL P, DEO M D, HANSON F V. Crude oil
characterization and regular solution approach to thermody-
namic modeling of solid precipitation at low pressure. Petro-
leum Science & Technology, 2004, 22(7/8): 973–990.
[22] MATZAIN A. Multiphase flow wax deposition modeling.
Tulsa: University of Tulsa, 1999.
[23] ZHANG Yu. Study on wax deposition in multiphase flow.
Beijing: China University of Petroleum (Beijing), 2011.
[24] DUAN J, LIU H, JIANG J, et al. Numerical prediction of wax
deposition in oil-gas stratified pipe flow. International Journal
of Heat & Mass Transfer, 2017, 105: 279–289.
[25] CHEN Dechun, LIU Junrong, WU Xiaodong, et al. Models for
predicting paraffin deposition profile in waxy oil wellbore.
Journal of the University of Petroleum, China (Edition of
Natural Science), 1999, 23(4): 36–38.
[26] SINGH P, WALKER J A, LEE H S, et al. An application of
vacuum insulation tubing for wax control in an arctic envi-
ronment. SPE Drilling & Completion, 2007, 22(2): 127–136.
[27] GAO Yonghai. Study on multi-phase flow in wellbore and
well control in deep water drilling. Dongying: China Univer-
sity of Petroleum, 2007.
[28] SUN Baojiang. Multiphase flow in oil and gas engineering.
Dongying: China University of Petroleum Press, 2013.
[29] BENDIKSEN K, MAINES D, MOE R, et al. The dynamic
two-fluid model OLGA: Theory and application. SPE Pro-
duction Engineering, 1991, 6(6): 171–180.
[30] RØNNINGSEN H P, SØMME B, PEDERSEN K S. An im-
proved thermodynamic model for wax precipitation: Experi-
mental foundation and application. In: Proceedings of 8th Inter-
national Conference on Multiphase '97. Cannes, France: Me-
chanical Engineering Publications, 1997.