Applied Thermal Engineering 207 (2022) 118122

Contents lists available at ScienceDirect

Applied Thermal Engineering

journal homepage: [Link]/locate/apthermeng

Research Paper

Experimental study on thermal hydraulic characteristics of natural

circulation loop under motion condition
Zhixian Lai, Wenxi Tian *, Chong Chen, Mingjun Wang, Kui Zhang, Suizheng Qiu, Guanghui Su
State Key Laboratory of Multiphase Flow in Power Engineering, School of Nuclear Science and Technology, Xi’an Jiaotong University, Xi’an 710049, China

A R T I C L E I N F O A B S T R A C T

Keywords: In this present paper, an experimental research was carried out to investigate the influence of various motion
Experiment study conditions on the flow and heat transfer characteristics in natural circulation loops. A natural circulation loop
Motion conditions was built on the six-degree-of-freedom motion experiment platform at Xi’an Jiaotong University (XJTU). First, it
Natural circulation
was found that the small acceleration heaving would not cause significant changes on flow rate of the natural
Flow and heat transfer
circulation loop. Then, the influence mechanism of rolling motion on the natural circulation loop is different as
the central axis of rolling is changed. The increase of additional inertial force along the coolant flow direction
causes larger flow fluctuation. The periodic average frictional resistance coefficient under different working
conditions is obtained, which is divided into flow fluctuation region (Re < 5000) and turbulence region (5000 <
Re < 35000). Last, according to the experimental data, a mathematical model to predict the periodic average
frictional resistance coefficient which is suitable for the turbulent region is obtained, and the deviation between
the predicted value and the experimental data is less than 20%. Beyond that, the period average Nu under
different working conditions is also achieved. It shows that the rolling motion promotes the heat transfer with the
small flow rate, while the rolling motion affects the heat transfer characteristics mainly by affecting the flow rate
under the large flow rate conditions. Based on our research, some suggestions on the layout of marine natural
circulation loop were given.

industry and ship design.

1. Introduction Tan et al. [12,13] carried out a series of experimental studies on the
heat transfer characteristics of single-phase flow and the instability of
Under the influence of wind and waves, floating platforms and ships two-phase flow in the natural circulation loop under rolling conditions
at sea will have complex motions of six degrees of freedom, as illustrated and some empirical equations of heat transfer coefficient were obtained.
in Fig. 1, include roll, pitch, yaw (rotational motions) and surge, sway, The experimental and theoretical researches on “Mustu”, NSR-7 and
heave (linear motions) about the x-, y- and z-axis. Motion conditions other Marine reactors were carried out to analyze the influence of ship
cause the obvious change on flow and heat transfer characteristics of movement on natural circulation characteristics [14–17]. Kim et al. [18]
marine applications such as turbines [1] and photovoltaic [2]. In recent built a miniature modulated natural circulation experimental facility
years, great effort has been taken to research on the mechanism of this based on the integrated modular small reactor SMART. Through ex­
change [3,4]. As a kind of reliable energy conversion and transfer sys­ periments and one-dimensional program simulation, they found that the
tem, natural circulation loop has long been widely applied in modern natural circulation loop presented different flow characteristics at
engineering fields, such as solar thermal systems [5], nuclear applica­ different tilt angles. Zahraddeen et al. [19] built an experimental double
tions [6], geothermal energy extraction [7], electronic devices cooling loop and successfully established natural circulation under different
[8], refrigeration system [9] and ship propulsion [10]. Due to the low pressure. Hwang et al. [20] carried out an experimental study on the
driven head and the influence of additional force under the ocean mo­ critical heat flux under rolling conditions and investigate the variation of
tion conditions, the thermal hydraulic characteristics in a natural cir­ critical heat flux of R134a in the upward flow along the uniform heating
culation loop are significantly varied compared with those under the channel. Vijayan et al. [21] explored the influence of cold and heat
static conditions [11]. Therefore, the natural circulation under ocean source locations on the single-phase natural circulation characteristics
conditions has been a research hotspot in the field of offshore energy experimentally. Wang et al. [22] carried out experiments on the natural

* Corresponding author.
E-mail address: wxtian@[Link] (W. Tian).

[Link]
Received 3 October 2021; Received in revised form 15 December 2021; Accepted 17 January 2022
Available online 20 January 2022
1359-4311/© 2022 Published by Elsevier Ltd.
Z. Lai et al. Applied Thermal Engineering 207 (2022) 118122

Nomenclature Tf Periodic mean fluid temperature, K

Tw Periodic mean wall temperature, K
cp Specific heat at constant pressure, kJ⋅kg− 1 T f,in Periodic average temperature at the inlet of heating tube, K
Din Inner diameter of heating tube, m
T f,out Periodic average temperature at the outlet of heating tube,
Do Outer diameter of heating tube, m
K
f Modified frictional resistance coefficient
T wi,in Periodic average inner wall temperature of measuring
fL Frictional resistance coefficient
point, K
g Gravitational acceleration, m⋅s− 2
h Heat transfer coefficient, W⋅m− 2⋅℃-1 T w,out Periodic average outer wall temperature of measuring
k, n Coefficient point, K
L Length of heating tube, m u Mean flow velocity, m⋅s− 1
Δptest Pressure drop gauge reading, Pa W Periodic average mass flowrate, kg⋅s− 1
Δptemp Difference of gravity pressure drop caused by temperature θ Tilt angle of heating tube, rad
difference, Pa θm Maximum tilt angle of heating tube, rad
δ(Δpadd ) Difference of additional pressure drop, Pa ρ Density, kg⋅m− 3
Δpf Frictional pressure drop, Pa ρf Average density of fluid in heating tube, kg⋅m− 3
q Periodic average heat flux, W⋅m− 2 ρl Fluid density in impulse line, kg⋅m− 3
qw Periodic mean average heat flux, W⋅m− 2 λ Thermal conductivity of coolant, W⋅m− 1⋅K− 1
qin Periodic average heat flux on the inner wall, W⋅m− 2 λw Wall thermal conductivity at outer wall temperature,
R Uncertainty of a certain experimental measurement W⋅m− 1⋅K− 1
quantity μw Dynamic viscosity at wall temperature, Pa⋅s
Rdau Uncertainty of data acquisition system μf Dynamic viscosity at mean fluid temperature, Pacs
Rmuu Uncertainty of measuring instrument Nu Nusselt number
t Time Re Reynolds number
T Rolling period

bubble rising behavior under the ocean motion conditions.

The existing experimental research mainly focus on the effects of
simple motion conditions such as heaving and rolling on the natural
circulation characteristics, while the actual ocean motion conditions
faced by ships are complex motion with six degrees of freedom. Obvi­
ously, more valuable experimental data are required to validate the
system code and simulation methods under more complex motion con­
ditions. The primary goal of this paper is to find out the influence
mechanism of complex ocean conditions on the characteristics of natural
circulation, give suggestions to the layout of marine natural circulation
loop and provide key references and appendices for numerical analyses.
A six-degree-of-freedom ocean motion condition simulation platform
was built and the characteristics of natural circulation loop under
various ocean motion conditions were studied experimentally.

2. Experimental apparatus and method

Fig. 1. Diagram of 6 degree of freedoms motion. 2.1. Experimental apparatus

circulation loop of narrow rectangular channels under rolling condi­ In order to study the flow and heat transfer characteristics of the
tions. Tang et al. [23] studied the two-phase flow instability in a parallel natural circulation loop under typical ocean conditions, the corre­
rectangular channel under rolling conditions. The existing experimental sponding experimental system was built in this study, as shown in Fig. 2.
studies mainly focus on the influence of simple motion conditions such The experimental apparatus includes the main loop, cooling water loop,
as the undulation and rolling on natural circulation characteristics, ocean motion simulation platform, electric heating system and pressure
while the actual ocean motion conditions are extremely complex with stabilizing system. The main loop is a closed natural circulation loop,
six degrees of freedom motion. At present, there are few experimental including the serpentine preheater, heating tube, casing condenser and
studies on the influence of complex ocean conditions on natural circu­ other auxiliary equipment.
lation characteristics. The overall height of loop is 3.8 m and the width is 1.4 m. The
In terms of numerical simulation using system code and CFD method, heating tube is connected to the main loop through an insulating flange.
series of simple motion condition cases have been studied in the liter­ The inner diameter of tube is 14 mm and the length is 1.2 m. Inner
ature [24]. Yan et al. [25] carried out an experimental study on the diameter of other pipe segments of the main loop is 19 mm. The tube
natural circulation under rolling conditions and improved the RELAP5/ used in our research was made of chemically polished 316L stainless
MOD3.3 program for the motion conditions simulation. Tian et al. [26] steel, its roughness was under 0.2 μm. The working fluid in the loop is
developed an analysis program based on SIMPLE algorithm, analyzed water. The overall arrangement of the experimental apparatus is shown
the flow instability of natural circulation loop of parallel channel under in Fig. 3. The serpentine preheater is located at the bottom of the main
typical ocean motion condition and obtained the corresponding flow loop. Its outlet is connected with the inlet of heating tube, and the casing
instability boundary. Chen et al. [27] used CFD method to study the condenser is located at the top of main loop. The cooling water loop is

2
Z. Lai et al. Applied Thermal Engineering 207 (2022) 118122

Fig. 2. Experimental apparatus of the natural circulation loop under ocean condition.

Fig. 3. Main loop diagram.

connected to the casing condenser through the hose, and the pressure
stabilizing system is connected to the inlet of serpentine preheater
Table 1 through the hose. The preheater and heating tube are heated directly by
Experimental parameters of the natural circulation loop under electric connections with the clamping electrode. The main loop is
ocean motion condition. placed on the ocean motion condition simulation platform, which
Parameters Range adopts the Stewart structure and can realize the six-DOF motion con­
dition simulations.
Pressure, MPa 0.2–1.1
Flow rate, kg⋅m− 2⋅s− 1 100–667
Power of heating tube, kW⋅m− 2 10–180
Power of preheater, kW⋅m− 2 0–210 2.2. Experimental method and operating conditions
Maximum rotation angle, ◦ 5–20
Rotation period, s 4–16 Due to the additional Coriolis force caused by ocean motion, we use
Maximum heaving distance, m 0.1–0.3 electromagnetic flowmeter instead of Coriolis force in our research.
Heaving period, s 6–16
Therefore, water with constant conductivity was chosen as coolant.

3
Z. Lai et al. Applied Thermal Engineering 207 (2022) 118122

line. Wall temperature measuring points are arranged at equal intervals

of 100 mm on the heating tube. Because the simulation of ocean motion
conditions introduces additional inertial forces, the electromagnetic
flowmeter is used to measure the flow rate of main loop. In addition, the
pressure difference between the inlet and outlet of heating tube and the
pressure at the inlet of heating section are also measured. The mea­
surement instruments used in the experiment includes: Rosemount
1151AP pressure transmitter, Rosemount 1151HP differential pressure
transmitter, PXI1002-SCXI1000 data acquisition system produced by
National Instruments.
The layout of each measuring point on the heating tube is shown in
Fig. 4, in which four thermocouples are used to measure the wall tem­
perature at each wall temperature measuring point and the average
value is taken as the wall temperature at that point. The accuracy,
corresponding measurement deviation and uncertainty of various
measuring instruments used in the experiment are shown in Table 2. The
uncertainty of measurement can be expressed as the arithmetic root of
the sum of squares of all the data involved in data processing according
to Kline method [28]. Then the uncertainty of a certain experimental
measurement quantity is:
√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
( )2 ( )2
δR δRdau δRmuu
= + (2.1)
R Rdau Rmuu

3. Comparison of loop flow characteristics under different

motion conditions

With the influence of ocean wind and waves, ships will have complex
motions including six degrees of freedom [14], most of the working
Fig. 4. Measuring points in the heating tube. conditions can be decomposed into heaving along the Z axis and rotation
along the center axis of X axis and Y axis. The rotation motion along the
center axis of X axis and Y axis as rolling and pitching is defined as
Table 2
shown in Fig. 3. For the flow of coolant in the pipe under the condition of
Measurement deviation and uncertainty.
ocean movement, the heaving causes the flow rate and pressure drop in
Measured parameters Measurement Uncertainty the pipe to oscillate [29,30]. The additional radial acceleration leads to
deviation
secondary flow in the channel, while the axial acceleration makes the
Fluid temperature ±0.75% 0.75% coolant flow obviously oscillate. The effect of axial acceleration on the
Wall temperature 0.75%
±0.75%
coolant flow characteristics in the tube is stronger than that of radial
Pressure ±0.25% 0.251%
Pressure drop between inlet and outlet of ±0.25% 0.251%
acceleration [12]. Similarly, when the relative position and direction of
heating tube the additional inertial forces brought by the ocean motion conditions
Flow rate ±0.5% 0.5% and the natural circulation loop changes, the flow and heat transfer
Data signal ±0.02% – characteristics of the loop would vary significantly. In this paper, the
Length
flow characteristics of natural circulation loop under different motion
±0.01% –
Diameter ±0.1% –
Heating power ±0.05% 0.05% conditions are studied firstly.
Wall heat flux of the heating tube – 1.086% In this experiment, the temperature change of single-phase fluid at
Flow boiling heat transfer coefficient h – 1.324% the inlet and outlet is small, so the density of fluid can be regarded as a
Nu 1.324%
–
consistent value. Therefore, the frictional pressure drop at the inlet and
outlet of heating pipe can be expressed as:
Before the start of experiment, the non-condensable gas in the main loop Δpf = Δptest + Δptemp − δ(Δpadd ) (3.1)
and the pressure lead pipe is discharged through the exhaust valve first
to insure the loop is full of coolant. And the pressure in the loop is raised where Δptest /Pa is the pressure drop of gauge reading number; Δptemp /Pa
to the specified working condition, make sure the main loop is a single- is the difference of gravity pressure drop caused by the temperature
phase buoyancy-driving loop. Through adjusting the heating power of difference between the impulse line and heating tube; δ(Δpadd )/Pa is the
the preheater, heating tube, and the flow of cooling water, the loop difference of additional pressure drop between the impulse line and the
forms a stable natural circulation under the static state. Then the ocean heating tube.
motion simulation platform is started and the experimental data are
recorded within 10 rolling periods. The parameters of various operating Δptemp = (ρl − ρf )gLcosθ (3.2)
conditions in the experiment are shown in Table 1.
where ρl /kg∙m− 3 is the fluid density in the impulse line; ρf /kg∙m− 3 is
the average density of fluid in the heating tube; θ/◦ is the tilt angle of the
2.3. Parameter measurement and error analysis heating tube.
δ(Δpadd ) is the difference of the additional pressure drop between the
In this experiment, K-type armored thermocouple produced by heating tube and impulse line due to the under motion conditions, which
Shanghai Instrument Factory is used to measure the fluid and wall is related to the relative position of the impulse line and the heating
temperature. The measuring points include the inlet and outlet of pre­ tube. In this experiment, the additional pressure difference under
heater, heating pipe and condenser, the fluid temperature of impulse

4
Z. Lai et al. Applied Thermal Engineering 207 (2022) 118122

Fig. 5. Flow characteristics of the main loop under rolling motion.

Fig. 6. Flow rate and friction pressure drop under the heaving condition.

Fig. 7. Flow rate and frictional pressure drop under pitching condition.

various motion conditions is directly measured. Fig. 6 shows the variation of the friction pressure drop and flow rate
Fig. 5 shows the relationship between the frictional pressure drop at in the heating tube with time under different heaving conditions. It can
the inlet and outlet of the heating tube and the flow rate of the main loop be seen that the flow rate and the frictional pressure drop in the heating
with time under different rolling conditions. Rolling motion causes pe­ section change with a small amplitude compared to the rolling condi­
riodic oscillation of the friction pressure drop and the flow rate in the tions without obvious periodicity. This is because the direction of
main loop, and the oscillation period is consistent with the rolling period additional acceleration is parallel to the direction of the gravitational
with a certain phase difference. The oscillation amplitude of friction acceleration under heaving conditions, which makes the difference of
pressure drop and flow rate increases with the increase of angular ac­ coolant gravity pressure drop in the cold and hot section of the natural
celeration of rolling motion. circulation loop change. But the amplitude of additional acceleration is

5
Z. Lai et al. Applied Thermal Engineering 207 (2022) 118122

Fig. 8. Variation of the main loop flow rate with different rotation axis deflection angle.

far less than the acceleration of gravity. changes obviously differently under the rolling and pitching conditions.
Fig. 7 shows the variation of the friction pressure drop and flow rate Fig. 8 indicates the flow rate variation of the natural circulation loop
in the heating tube with time under pitching conditions. As can be seen with the same rotation angle and different rotation axis deflection an­
from Fig. 7(a), the friction pressure drop in the heating tube also pre­ gles. With the decrease of the angle between rotation axis and Y axis, the
sents periodic changes under pitching conditions, but the oscillation amplitude of flow oscillation increases. When the rotation axis is
amplitude is much smaller than that under the rolling conditions with recombined with the Y axis, the flow oscillation amplitude is the
the same parameter as shown in Fig. 5, while the flow rate under maximum, and when the rotation axis is recombined with the X axis, the
pitching condition does not show obvious periodic changes. Under the flow oscillation amplitude is the minimum. In the natural circulation
condition of ocean movement, there are two main reasons for the system under ocean conditions, the flow fluctuation mainly depends on
oscillation of frictional pressure drop. The first is the change of frictional the relative relationship among the additional force, driving force and
resistance caused by the change of flow velocity during the rotation resistance. The additional force is decomposed into two components,
process. The second is the secondary flow caused by the rotation mo­ which are the parallel to and perpendicular to the direction of coolant
tions, which destroys the original boundary layer structure, enhances flow. The former will bring additional pressure drop to the coolant and
the turbulence intensity in the tube, and leads to the change of friction directly cause coolant flow oscillation, as shown in Eq. (3.3), where the
resistance coefficient. The former is mainly affected by the additional four terms on the right side of the equation are driven pressure drop,
forces parallel to the flow direction, and the oscillation amplitude of the gravity pressure drop, friction pressure drop and additional pressure
friction resistance caused by the former is large, while the latter is drop. As the driven pressure drop and frictional pressure drop of natural
mainly affected by the additional forces perpendicular to the flow di­ circulation are smaller than the additional pressure drop, the additional
rection. In this study, the overall structure of the main loop is an pressure drop plays a major role in the flow oscillation. Under the same
approximate rectangular circuit. Under the pitching conditions, most of rotation angle, the magnitude of additional force remains unchanged.
the pipe segments are subjected to the additional force perpendicular to When the angle between the rotation axis and the Y axis decreases, the
the flow direction, while under the rolling conditions, most of the pipe component of the additional force parallel to coolant flow direction
segments are subjected to the additional force parallel to the flow di­ increases, resulting in an increase in the additional pressure drop and
rection. This is the main reason that the friction coefficient of the loop more severe flow oscillation.
∑ dui ∑ ∑ ∑ ∑
ρi li = Δpd + Δpg + Δpf + Δpa (3.3)
dt
Through the above analysis, it can be seen that among all the single-
degree of freedom motions, the rolling motion has the greatest influence
on the characteristics of the natural circulation loop. It will cause a large
oscillation of the coolant flow in the natural circulation loop, and the
coolant flow will directly affect the heat removal capacity of the natural
circulation loop. Since during most time of the ship’s voyage, the bow is
often directed perpendicular to the waves, rolling is the most frequent
condition. Therefore, in avoid of large oscillation of coolant flow, nat­
ural circulation loops were suggested install in a way that the X axis of
loop is perpendicular to the X axis of ship. In this way, when the ship is
rolling, the loop is pitching. Moreover, the flow and heat transfer
characteristics of the natural circulation circuit under the rolling con­
dition are studied in depth in the following sections.

4. Thermal hydraulic characteristics under rolling condition

4.1. Flow characteristics

Fig. 9. Periodic average friction resistance coefficient under different oper­ According to the frictional pressure drop measured in the experi­
ating condition. ment, the frictional resistance coefficient can be obtained through Darcy

6
Z. Lai et al. Applied Thermal Engineering 207 (2022) 118122

2π t
θ = θm sin( )
T

d2 θ 4π 2 2π t
∝= = − θm sin( ) (4.4)
dt2 T2 T
As we discussed in Section 3, the additional force parallel to the di­
rection of coolant flow will bring additional pressure drop to the coolant
and directly cause coolant flow oscillation. And the additional force
increases as the angular acceleration increase. Since we dealt with the
data in 10 periods averagely, we added the correction term in form of
1 + (sinθm )n to Blasius formula, result in Eq. (4.3).
Through nonlinear curve fitting, the corresponding correction coef­
ficient can be obtained: n = 1.3, k = 0.3236, and the prediction equation
can be obtained:

0.3236[1 + (sinθm )1.3 ]

f = (4.5)
Re0.25

Fig. 10. Results of the average frictional resistance coefficient predic­

Application scope of the prediction formula is: 5000 < Re < 35000,
tion formula. 5◦ < rolling angle < 15◦ , 8 < rolling period < 16 s.
Fig. 10 illustrates the results of the periodic average frictional
resistance coefficient prediction formula under ocean motion conditions
formula:
established in this study. More than 97.5% of experimental data fit the
2Δpf Din prediction formula with an error less than ± 20%.
fL = (4.1)
ρLu2

where Din /m is inner diameter of heating tube; u is mean flow velocity; 4.2. Heat transfer characteristics
L/m is length of heating tube; ρ/kg∙m− 3 is fluid density.
In the processing of experimental data, it is assumed that the heating During the experiment, the flow rate in the main loop will change
tube is isothermal, which is slightly different from the actual working periodically, so the heat transfer process is a non-equilibrium process. In
conditions. Therefore, the results are modified in the same way as the this study, the periodic average heat transfer characteristics under ocean
Sieder-Tate formula: motion conditions are mainly studied. In other words, the average wall
μw 0.6 temperature, average fluid temperature, average flow rate and average
f = fL Â⋅( ) (4.2) heat flux in several periods are taken, and the method similar to steady
μf
state is adopted based on the obtained average parameters.
where μw , μf / Pa⋅s are hydrodynamic viscosity based on the wall tem­ The heat transfer coefficient in the heating tube is:
perature and mean fluid temperature respectively. qw
After the periodical average temperature of heating tube wall and h= (4.6)
Tw− Tf
coolant at the inlet was stable, the average velocity and frictional
pressure drop of the heating tube in 10 periods were calculated, and the Then Nu number can be expressed as:
average frictional resistance coefficient of the single tube channel in the Din
natural circulation loop under ocean motion conditions was obtained by Nu = h (4.7)
λ
using Eqs. (4.1) and (4.2). Fig. 9 indicates the characteristics of the
frictional resistance coefficient of the main loop under static and where qw /W∙m− 2 is periodic mean heat flux; Tw /K is periodic mean wall
different rolling conditions, which are consistent with the variation temperature; Tf /K is periodic mean fluid temperature; Din /m is diameter
trend of the traditional frictional resistance coefficient in each Re region. of heating tube; λ/W∙m− 1∙K− 1 is the thermal conductivity of coolant.
But it should be pointed out that when Re is less than 5000, although the For single-phase flow heat transfer, the fluid temperature is taken as
frictional resistance coefficient is in good agreement with the calculation the average temperature at the inlet and outlet, i.e.:
relationship of the traditional laminar flow area, the flow oscillation
range in the main loop is large, and even the reverse flow phenomenon T f = (T f ,in + T f ,out )/2 (4.8)
will appear. This region is defined as the flow fluctuation region. Under
For the wall temperature, taken as the average temperature of 11
the influence of rolling motions, the average frictional resistance coef­
measuring points along the length direction:
ficient of the period in the turbulent region is higher than that in the
static conditions. This is because the lateral mixing of fluid is enhanced ∑11

by the rolling motions, which intensifies the scouring effect of fluid on Tw = ( T wi,in )/11 (4.9)
the wall and accelerates the energy dissipation near the wall.
i=1

Based on the experimental data obtained in this study, using the form Since only the temperature of the outer wall was measured in the
of Blasius resistance relation and adding the resistance correction co­ experiment, according to the thermal conductivity differential equation,
efficient under the ocean motion conditions, the prediction formula of the temperature of the inner wall can be obtained:
periodic average frictional resistance coefficient under ocean motion qin Din
conditions is established in the form of: T w,in = T w,out − ln(Do /Din ) (4.10)
2λw
n
k[1 + (sinθm ) ]
f = (4.3) where qin is periodic average heat flux on the inner wall, λw /W∙m− 1∙K− 1
Re0.25
is wall thermal conductivity at outer wall temperature.
Considering typical rolling condition, the rolling angle and angular The heating tube adopts 316L stainless steel, and its thermal con­
acceleration can be described as follow: ductivity is:

7
Z. Lai et al. Applied Thermal Engineering 207 (2022) 118122

rolling motion mainly affects the heat transfer characteristics of the

heating tube by affecting the natural circulation flow of the main loop.
Similar previous Nu predict formula [12] was compared in Fig. 11. It is
shown that the formula can only apply on severe motion condition,
when the angular acceleration is small, the agreement between experi­
ment value and calculated value is poor.
Fig. 12 illustrates the temperature variation of the wall surface of
heating tube and the inlet and outlet fluid caused by rolling motions
under different Re conditions. The period average frictional resistance
coefficient increases due to the rolling motion. The period average flow
rate under rolling is lower than that under the static condition, and the
inlet and outlet temperatures of heating tube under rolling condition are
slightly higher than that under the static condition. When Re is small (Re
= 977), it is in the heat transfer enhancement region as shown in Fig. 10.
The wall temperature of measurement point TW1, which is closer to the
inlet, is greatly affected by the inlet fluid temperature and still shows an
oscillating upward trend. However, the wall temperatures of measure­
ment points TW4, TW7 and TW11, which are far from the inlet, all show an
oscillating downward trend, because the rolling motion increases the
turbulence intensity of coolant flow and enhances the heat transfer be­
Fig. 11. Period average Nu under different operating condition. tween the wall and the fluid. When Re is large (Re = 10722), the
changing trend of temperature on the wall is consistent with fluid
λw = 14.85 + 0.014337Tw,out (4.11) temperature at the inlet and outlet of heating tube, and the temperature
changing trend is the same as that of flow rate in the main loop.
For the periodic average heat flux, it is calculated according to the
Through above analyses, for natural circulation loops operate in heat
periodic average fluid inlet and outlet temperature through the heat
transfer enhancement region with small flow rate and Re, the suggested
balance method:
installing way is that the X axis of loop parallel to X axis of ship. In this
q = Wcp (T f ,out − T f ,in )/πDin L (4.12) way, the heat transfer enhancement brought by rolling motion can
reduce the wall temperature of heating tube.
Fig. 11 illustrates the variation of periodic mean Nu under static and
different rolling conditions. When the flow rate is small (Re < 5000), the 5. Conclusion
rolling condition has a certain promotion effect on the wall heat transfer,
because the rolling motion strengthens the turbulence intensity near the In this paper, a natural circulation loop was built on a six-degree-of-
wall. With the increase of velocity and Re (5000 < Re < 35000), the freedom motion simulation platform. First, the flow characteristics in
turbulence intensity of the flow itself increases, and the turbulence the loop under different motion conditions were studied, and then the
enhancement caused by rolling becomes insignificant. Nu does not flow and heat transfer characteristics of the natural circulation loop
change obviously compared to the static conditions. At this region, the

Fig. 12. Temperature variation of the wall surface and the inlet and outlet of heating tube with different Re.

8
Z. Lai et al. Applied Thermal Engineering 207 (2022) 118122

under rolling condition were studied in depth. To the best of the authors’ [5] D.J. Close, The performance of solar water heaters with natural circulation,
Retrieved from, Sol. Energy 6 (1) (1962) 33–40, [Link]
knowledge, the following contributions/ findings are novel in this work:
ience/article/pii/0038092X62900968, [Link]
90096-8.
(1) The influence mechanism of different ocean motion conditions on [6] D.N. Basu, S. Bhattacharyya, P.K. Das, A review of modern advances in analyses
the characteristics of natural circulation loop are different. Spe­ and applications of single-phase natural circulation loop in nuclear thermal
hydraulics, Nucl. Eng. Des. 280 (2014) 326–348.
cifically, heaving with small acceleration will not cause signifi­ [7] D.B. Kreitlow, G.M. Reistad, C.R. Miles, G.G. Culver, Thermosyphon models for
cant changes in the flow characteristics of the natural circulation downhole heat exchanger applications in shallow geothermal systems, J. Heat
loop. The additional force is decomposed into two components, Transfer 100(4) (1978) 713–719. Retrieved from [Link]
1.3450883. doi:10.1115/1.3450883.
which are parallel to and perpendicular to the direction of [8] P.E. Tuma, H.R. Mortazavi, Indirect thermosyphons for cooling electronic devices,
coolant flow, respectively. The former brings additional pressure Electronic Cooling 12 (2006) 26–32.
drop to the coolant and directly causes the fluctuation of coolant [9] A. Hu, L. Zou, The progress and research on enclosed refrigeration system with
thermo-siphon and self-injection, Cryogenics Superconductivity (2011).
flow. The flow fluctuation enlarges with the increase of addi­ [10] D. Lu, Z. Xiao, B. Chen, A new method to derive one set of scaling criteria for
tional inertial force along the coolant flow direction. Under reactor natural circulation at single and two-phase conditions, Retrieved from,
rolling condition, the flow and heat transfer characteristics of the Nucl. Eng. Des. 240 (11) (2010) 3851–3861, [Link]
ence/article/pii/S0029549310005261, [Link]
natural circulation loop vary significantly. nucengdes.2010.08.012.
(2) The periodic average frictional resistance coefficient of the nat­ [11] G. Su, J. Zhang, Y. Guo, S. Qiu, Z. Yu, D. Jia, Effects of ocean conditions upon the
ural circulation loop under different operating conditions was passive reisidual heat removal system (PRHRS) of ship reactor. Atomic Energyence
& Technology.
investigated, which was divided into flow fluctuation region (Re
[12] S.-C. Tan, G.H. Su, P.-Z. Gao, Experimental and theoretical study on single-phase
< 5000) and turbulent region (5000 < Re < 35000). In the flow natural circulation flow and heat transfer under rolling motion condition, Appl.
fluctuation region, coolant reverse flow occurs in the natural Therm. Eng. 29 (14-15) (2009) 3160–3168.
circulation loop. A prediction formula for the periodic average [13] S. Tan, Z. Wang, C. Wang, S. Lan, Flow fluctuations and flow friction characteristics
of vertical narrow rectangular channel under rolling motion conditions, Retrieved
frictional drag coefficient suitable for the turbulent region was from, Exp. Therm Fluid Sci. 50 (2013) 69–78, [Link]
obtained, and the deviation between the predicted value and the /[Link]?eid=2-s2.0-84880741983&doi=10.1016%[Link].2013
experimental data is less than 20%. .05.006&partnerID=40&md5=bac6053a41f74a6256cdaaf2109300dd, https://
[Link]/10.1016/[Link].2013.05.006.
(3) The period average Nu of the natural circulation loop under [14] I. Ishida, T. Kusunoki, H. Murata, T. Yokomura, M. Kobayashi, H. Nariai, Thermal-
different operating conditions was achieved. It is found that when hydraulic behavior of a marine reactor during oscillations, Nucl. Eng. Des. 120 (2-
the flow velocity is small, the rolling motion enhances the tur­ 3) (1990) 213–225.
[15] T. Ishida, T. Yoritsune, Effects of ship motions on natural circulation of deep sea
bulence intensity near the wall and promotes the heat transfer to research reactor DRX, Nucl. Eng. Des. 215 (1-2) (2002) 51–67.
a certain extent. Experiments show that the faster the flow ve­ [16] H. Murata, I. Iyori, M. Kobayashi, Natural circulation characteristics of a marine
locity is, the greater the turbulence intensity of coolant flow be­ reactor in rolling motion, Retrieved from, Nucl. Eng. Des. 118 (2) (1990) 141–154,
[Link] https://
comes (5000 < Re < 35000) and the rolling motion mainly affects [Link]/10.1016/0029-5493(90)90053-Z.
the heat transfer characteristics by affecting the flow rate of [17] H. Murata, K.I. Sawada, M. Kobayashi, Natural circulation characteristics of a
natural circulation loop. marine reactor in rolling motion and heat transfer in the core, Retrieved from,
Nucl. Eng. Des. 215 (1–2) (2002) 69–85, [Link]
(4) In avoid of large oscillation of coolant flow, natural circulation
uri?eid=2-s2.0-0036603869&doi=10.1016%2fS0029-5493%2802%2900042-0
loops were suggested install in a way that the X axis of loop is &partnerID=40&md5=7d3344dd20d191dbec367d47b4bb9df4, [Link]
perpendicular to the X axis of ship. For natural circulation loops 10.1016/S0029-5493(02)00042-0.
operate in heat transfer enhancement region with small flow rate [18] J.-H. Kim, T.-W. Kim, S.-M. Lee, G.-C. Park, Study on the natural circulation
characteristics of the integral type reactor for vertical and inclined conditions,
and Re, the suggested installing way is that the X axis of loop Nucl. Eng. Des. 207 (1) (2001) 21–31.
parallel to X axis of ship. [19] A. Zahraddeen, J. Chen, K. Cheng, F. Zhao, S. Tan, Preliminary experimental
validation of multi-loop natural circulation model based on RELAP5/SCDAPSIM/
MOD 4.0, Retrieved from, Int. J. Adv. Nucl. Reactor Des. Technol. 2 (2020) 25–33,
[Link]
Declaration of Competing Interest
[Link]
[20] J.-S. Hwang, Y.-G. Lee, G.-C. Park, Characteristics of critical heat flux under rolling
The authors declare that they have no known competing financial condition for flow boiling in vertical tube, Retrieved from, Nucl. Eng. Des. 252
interests or personal relationships that could have appeared to influence (2012) 153–162, [Link]
312003512, [Link]
the work reported in this paper. [21] P.K. Vijayan, M. Sharma, D. Saha, Steady state and stability characteristics of
single-phase natural circulation in a rectangular loop with different heater and
cooler orientations, Retrieved from, Exp. Therm Fluid Sci. 31 (8) (2007) 925–945,
Acknowledgements
[Link] https://
[Link]/10.1016/[Link].2006.10.003.
This research has been supported by the National Natural Science [22] J. Wang, P. Gao, C. Chen, Experimental Study on Natural Circulation Flow in
Rectangular Channel Under Rolling, Singapore, 2017.
Foundation of China (Grant Nos. 11705139), (Grant Nos. 11622541).
[23] Y. Tang, B.-D. Chen, W.-Y. Xiong, X.-Z. Liu, Comparison of flow instabilities under
static condition and marine motion conditions based on experiments, Retrieved
References from, Ann. Nucl. Energy 70 (2014) 11–20, [Link]
e/article/pii/S0306454914000802, [Link]
anucene.2014.02.010.
[1] V.I. Khrisanov, B.F. Dmitriev, Marine electrical power industry with renewable
[24] M. Wang, Y. Wang, W. Tian, S. Qiu, G.H. Su, Recent progress of CFD applications in
energy carriers. Part 1. Wind and wave turbines of offshore power plants, Russian
PWR thermal hydraulics study and future directions, Ann. Nucl. Energy 150 (2021)
Electr. Eng. 87 (7) (2016) 409–416.
107836, [Link]
[2] H. Liu, Q. Zhang, X. Qi, Y. Han, F. Lu, Estimation of PV output power in moving
[25] B.H. Yan, L. Yu, The experimental and theoretical analysis of a natural circulation
and rocking hybrid energy marine ships, Retrieved from, Appl. Energy 204 (2017)
system in rolling motion, Prog. Nucl. Energy 54 (1) (2012) 123–131.
362–372, [Link]
[26] W. Tian, M. Gui, D. Wu, R. Chen, G.H. Su, S. Qiu, Study on flow instability in
730884X, [Link]
natural circulation loop with parallel channels under moving condition, Retrieved
[3] S.W. Chang, T.-M. Liou, J.S. Liou, K.-T. Chen, Turbulent heat transfer in a tube
from, Nucl. Eng. Des. 379 (2021), 111246, [Link]
fitted with serrated twist tape under rolling and pitching environments with
e/article/pii/S0029549321001989, [Link]
applications to shipping machineries, Retrieved from, Ocean Eng. 35 (16) (2008)
nucengdes.2021.111246.
1569–1577, [Link]
[27] C. Chen, M. Wang, X. Zhao, H. Ju, X. Wang, W. Tian, G. Su, Numerical study on the
789, [Link]
single bubble rising behaviors under rolling conditions, Retrieved from, Nucl. Eng.
[4] Y. Gu, Y. Ju, Investigation on the periodically oscillating pressure characteristics of
Des. 349 (2019) 183–192, [Link]
the flow in the rolling pipe, Retrieved from, Ocean Eng. 55 (2012) 1–9, https://
S0029549318311403, [Link]
[Link]/science/article/pii/S0029801812002594, [Link]
org/10.1016/[Link].2012.06.035.

9
Z. Lai et al. Applied Thermal Engineering 207 (2022) 118122

[28] S.J. Kline, F.A. Mcclintock, Describing Uncertainties in Single-Sample Experiments, 238 (1) (2008) 178–187, [Link]
Mech. Eng. (New York, N.Y.: 1919), 75(1) (1953). S0029549307004104, [Link]
[29] R. Pendyala, S. Jayanti, A.R. Balakrishnan, Flow and pressure drop fluctuations in a [30] R. Pendyala, S. Jayanti, A.R. Balakrishnan, Convective heat transfer in single-phase
vertical tube subject to low frequency oscillations, Retrieved from, Nucl. Eng. Des. flow in a vertical tube subjected to axial low frequency oscillations, Heat Mass
Transf. 44 (7) (2008) 857–864.

10