вход по аккаунту



код для вставкиСкачать
Proceedings of the 2017 25th International Conference on Nuclear Engineering
July 2-6, 2017, Shanghai, China
Andrea Bersano
Politecnico di Torino
Turin, Italy
Cristina Bertani
Politecnico di Torino
Turin, Italy
Mario De Salve
Politecnico di Torino
Turin, Italy
Nicolò Falcone
Politecnico di Torino
Turin, Italy
Within the field of research and development of innovative
nuclear reactors, in particular Generation IV reactors and Small
Modular Reactors (SMR), the design and the improvement of
safety systems play a crucial role. Among all the safety systems
high attention is dedicated to passive systems that do not need
external energy to operate, with a very high reliability also in the
case of station blackout, and which are largely used in
evolutionary technology reactors.
The aim of this work is the experimental and numerical analysis
of a passive system that operates in natural circulation in order
to study the mechanism and the efficiency of heat removal. The
final goal is the development of a methodology that can be used
to study this class of systems and to assess the thermal-hydraulic
code RELAP5 for these specific applications. Starting from a
commercial size system, which is the decay heat removal system
of the experimental lead cooled reactor ALFRED, an
experimental facility has been designed, built and tested with the
aim of studying natural circulation in passive systems for nuclear
applications. The facility has been simulated and optimized
using the thermal-hydraulic code RELAP5-3D. During the
experimental tests, temperatures and pressures are measured and
the experimental results are compared with the ones predicted by
the code.
The results show that the system operates effectively,
removing the given thermal power. The code can predict well the
experimental results but high attention must be dedicated to the
modeling of components where non-condensable gases are
present (condenser pool and surrounding ambient). This facility
will be also used to validate the scaling laws among systems that
operate in natural circulation.
Bruno Panella
Politecnico di Torino
Turin, Italy
External tube area [m2]
Tube outer diameter [m]
Tube inner diameter [m]
Radiation view factor wall-ambient
Grashof number
Gravity acceleration [m/s2]
Total heat transfer coefficient [W/m2K]
Convective heat transfer coefficient [W/m2K]
Equivalent radiative heat transfer coefficient
Thermal conductivity [W/mK]
Tube length [m]
Characteristic length [m]
Nusselt number
Prandtl number
Pressure [Pa]
Heat transfer rate wall-ambient [W]
Air temperature [K]
Film temperature [K]
Wall temperature [K]
Isothermal expansion coefficient [K-1]
Height difference [m]
Total emissivity
Kinematic viscosity [m2/s]
Density [kg/m3]
Stefan - Boltzmann constant [W/m2K4]
Heat flux [W/m2]
Copyright © 2017 ASME
Downloaded From: on 10/25/2017 Terms of Use:
The Decay Heat Removal Systems (DHR) is a fundamental
component in every nuclear power plant that allows the control
of the core temperatures both after a normal shutdown and when
an accident occurs. The growth of nuclear energy strongly
depends on the safety features implemented in the future
reactors. Because of that, when developing an innovative reactor
design, high attention is dedicated to safety features and, in
particular, to passive safety systems. The strength of this class of
systems is they do not need external energy to operate, resulting
in a higher reliability and safety level. This kind of systems are
widely used in advanced nuclear reactors as Generation IV ones
and Small Modular Reactors (SMR).
In this work, starting from a reference passive system
relying on natural circulation, an experimental facility has been
designed and built at Dipartimento Energia of Politecnico di
Torino, in order to study natural circulation in this kind of
systems both experimentally and using a thermal-hydraulic code.
The code used is RELAP5-3D, which is widely adopted to
design nuclear reactors components and to perform accidental
scenarios simulations.
The chosen reference system is the backup of the Decay
Heat removal system of the lead cooled fast reactor ALFRED,
which is the demonstrator of the European lead cooled reactor
prototype in the framework of the LEADER project [1].
ALFRED is a pool type reactor with a thermal power of 300
MWth (125 MWe) cooled by lead [2]. The reactor has four loops
of the Decay Heat Removal system (DHR1) connected to the
steam generators and four loops for the Decay Heat Removal
backup system (DHR2) immersed in the coolant pool. Every
DHR2 loop consists of a bundle of 80 lead-water bayonet heat
exchangers (called DIP cooler), an Isolation Condenser (IC)
consisting of a vertical tube bundle with two spherical plena and
the connecting piping [3].
Once the DHR2 is activated, it operates passively working
in natural circulation; it transfers the decay heat from the lead to
the Isolation Condenser pool, ensuring three days of operation
without the need of external actions. In order to study the natural
circulation regime in this category of systems, a scaled facility
has been designed taking inspiration from the DHR2 project. The
design of the experimental facility was supported by sensitivity
studies and pre-test calculations carried out by RELAP5-3D, in
order to determine the most appropriate operating conditions and
to foresee possible unstable flow patterns during the
experimental tests [4].
The facility replicates the key features of the DHR2 in a
simplified way, focusing mainly the attention on the bayonet heat
exchanger and the steam condenser. The thermal-hydraulic
behavior of the facility has been simulated using the code
RELAP5-3D, to study the natural circulation regime that occurs
and to validate the code for these particular applications.
the experimental facility contains just one bayonet heat
exchanger. Its cross section has been kept constant while the
length has been reduced to fit the maximum allowable height of
the laboratory, still maintaining a certain proportion in the
components length. Stainless steel AISI 304 piping NPS 1” SCH
160 has been chosen to allow the operation at high pressure but
in this paper only low pressure operations will be presented. The
bayonet heat exchanger is located in the lower part of the
experimental facility, while in the upper part there is the
condenser (Figure 1).
Water pool
Cold leg
Hot leg
Bayonet heat exchanger
Figure 1: Front view of the experimental facility
In the reference plant, every DHR2 system is composed of
80 bayonet heat exchangers, but due to power and space limits
The thermal power is provided to the system using two
highly insulated electric heating tapes that are wrapped around
the tube of the bayonet as shown in Figure 2. The heated length
Copyright © 2017 ASME
Downloaded From: on 10/25/2017 Terms of Use:
Table 1: Temperature and pressure measurement devices
is 1 m; using only one bayonet heat exchanger, the total power
involved is limited to approximately 1.5 kW. One of the greatest
differences between the reference system and the experimental
facility is the boundary condition at the DIP Cooler (bayonet heat
exchanger): for the experimental facility there is an imposed heat
flux condition on the outer wall of the bayonet while for the large
scale plant there is an imposed lead temperature condition. Due
to technical and economical limitations this is the best possible
solution for the time being. Since a relatively small power is
provided, a single vertical pipe is enough for the condenser,
which consists of two coaxial tubes without any plena. A large
water pool connected to the system allows the change of the
liquid water level in the condenser shell side, to modify the heat
transfer area and, therefore, to adjust the power transferred to the
Bayonet pressure
drop (Δp1)
Hot leg pressure
drop (Δp2)
Cold leg pressure
drop (Δp3)
Thermocouple K type
Rosemount 3051
absolute pressure
Rosemount 3051
differential pressure
Rosemount 3051
differential pressure
Rosemount 1151
differential pressure
±4 mbar
±0.05 mbar
±0.032 mbar
±0.014 mbar
Figure 2: Particular of the heating tapes coils on the outer bayonet wall
and of the thermocouple position
During the experimental tests, temperatures, pressures and
pressure drops are measured; in particular, temperatures are
measured at the inlet and outlet of every main component
(bayonet heat exchanger, condenser, hot and cold leg), whereas
absolute pressure is measured at the inlet of the bayonet and in
the bayonet inversion chamber, and pressure drops are measured
in the bayonet downcomer, in the hot leg and in the condenser
and cold leg; a schematic of the facility, together with the
measurement points, are reported in Figure 3. Labels from 1 to 8
refer to water side temperature and pressure; labels from A to C
refer to outer wall tube temperature.
The measurement devices type and the accuracy are
reported in Table 1.
The loop is completely filled with water at ambient
temperature except for the upper part of the bayonet annulus
(downstream the bayonet outlet) and the upper connection with
the safety valve where air remains trapped inside the loop. In
fact, at present, the facility design does not allow the complete
evacuation of the air from the previously mentioned two parts,
but this is not an issue to be addressed since these parts are in
positions that do not affect the circulation in the system;
furthermore, these parts containing air are useful since they
operate as expansion tanks, controlling the pressure increase in
the system.
Figure 3: Front view of the experimental facility with numbered
instrumentation nozzles
Copyright © 2017 ASME
Downloaded From: on 10/25/2017 Terms of Use:
Electric power is turned on after one minute from the
beginning of the test and it is kept constant for 5 hours. For the
first phase of the experimental tests the safety valve at the top of
the facility is set to open at 7 bars. Experimental data are
acquired every 0.1 s.
A series of preliminary experiments have been previously
performed in forced circulation with the insertion of a pump in
the facility in order to estimate which fraction of the total power
provided to the electric resistances actually reaches the coolant
inside the bayonet. The tests showed that an average power of
1.225 kW is received by the bayonet, which corresponds to
approximately 70% of the electric power supplied to the heating
throughout the present experimental test, the condenser (System
2 in Figure 4) is full of air, without a shell side water level as
typical of the DHR2 system. To obtain a good match between the
simulation and the experimental results a fine tuning of the
thermal losses is fundamental. This was achieved imposing an
estimated heat transfer coefficient on the outer face of the heat
transfer structures (Figure 4) and an ambient temperature; the
temperature value is the one measured in the laboratory ambient
during the experimental tests (20°C), while the heat transfer
coefficient has been calculated (both convection and thermal
radiation have been taken into account) and verified, after having
compared the simulation results with reference experimental
The experimental facility loop (System 1 in Figure 4) is
initially full of liquid water except for the pipe 232, upper part of
pipe 230 and upper part of annulus 214 that simulate the
components where trapped air is present.
The thermal-hydraulic behavior of the experimental facility
has been deeply analyzed using the RELAP5-3D code in order
to optimize the design (in particular the length of the condenser)
and the measurement devices (especially the differential pressure
Table 2 reports the correspondence between the components
of the RELAP5-3D input model and the different parts of the
experimental facility.
1 bar
System 1
System 2
Table 2: RELAP5 components correspondence with experimental
facility ones
Pipe 202
Pipe 204
Pipe 206
Pipe 208
Single vol 210
Annulus 212
Annulus 214
Pipe 216
Pipe 218
Pipe 220
Pipe 222
Pipe 224
Pipe 226
Time dep vol 228
Pipe 230
Pipe 232
Motor valve 233
Pipe 236
Pipe 238
Pipe 240
Heat Structures
HS 18
HS 12
Experimental facility correspondence
Cold leg
Bayonet inlet
Bayonet downcomer (non active part)
Bayonet downcomer (active part)
Bayonet inversion chamber
Bayonet riser annulus (active part)
Bayonet riser annulus (non active part)
Bayonet outlet
Hot leg
Hot leg – condenser connection
Condenser tube side inlet
Condenser tube side
Condenser tube side outlet
System 1 boundary conditions
Safety valve connection
Pneumatic valve connection
Safety valve
Connection pipe
Water filling pipe
Mass control valve and pipe
1 bar
HS 19
HS 10
For the first experimental campaign, the thermal insulation
was not installed; since the involved input thermal power is quite
low, thermal losses from bare components and pipes to the
ambient are sufficient to operate as final heat sink. Therefore,
SV 210
Figure 4: RELAP5-3D input model
Copyright © 2017 ASME
Downloaded From: on 10/25/2017 Terms of Use:
The total heat transfer coefficient is the sum of the two heat
transfer coefficients
Total heat transfer coefficient derivation
The calculation of a correct heat transfer coefficient is
fundamental to simulate properly the thermal losses to the
surrounding ambient that represents the heat sink of the system
in this configuration. The total heat transfer coefficient is the sum
of two terms: a convective heat transfer coefficient and an
equivalent radiative one.
The convective heat transfer coefficient is calculated
through the semi-empirical correlation for natural convection
over a vertical cylinder [5]
ℎ%&'( %
= 0.6  ∗ 

ℎN&N = ℎ%&'( + ℎL:M
The convective, the radiative and the total heat transfer
coefficients have been calculated for outer wall temperatures
from 22°C up to 234°C (using 20°C steps except for the first
12°C step), which correspond to air film temperature ranging
from 26.85°C to 126.85°C (300K – 400K). The air properties
have been evaluated accordingly to F.J. McQuillan et al. [6].
Figure 5 shows the results of the convective, radiative and total
heat transfer coefficient (considering air temperature to be fixed
at 20°C) as a function of the wall temperature. A table containing
the heat transfer coefficient values and the corresponding wall
temperatures has been implemented in the RELAP5 input in
order to simulate properly the heat losses based on the external
wall temperature of every component. The code automatically
interpolates the correct heat transfer coefficient as function of the
computed external wall temperature.
Eq. (1)
 8 − : ;%
Eq. (2)
All the properties are calculated at the film temperature, which
is the average between the air and the wall temperature
> =
8 + :
Eq. (8)
Eq. (3)
The characteristic length is calculated as
1 1
= +
%  D
Eq. (4)
Since in this case the thermal radiation cannot be considered
negligible, an equivalent heat transfer coefficient has been
calculated to take into account the radiative thermal losses. For
this calculation the surrounding ambient has been considered as
a black body and the emissivity of the weathered stainless steel
has been set to 0.85. The radiative thermal power transfer from
the pipe outer surface to the ambient can be written as
8→: =
 84 − :4
8→: 8
Figure 5: Convective, radiative and total heat transfer coefficient as a
function of the wall temperature
Eq. (5)
Since the system is surrounded by the ambient the view factor
8→: can be set equal to 1. Considering the heat flux it is possible
to rearrange the previous equation as
=  84 − :4
The initial conditions of the simulations (air and water
temperature and pressure) have been set accordingly to the
corresponding measured values; also, the power transient that
has been carried out is the same:
• t [0, 1 min): power source OFF
• t [1 min, 5 hours]: power source ON at 100%.
The power value for the RELAP5 simulations has been fixed to
1.225 kW that is the average computed power transferred to the
bayonet in a series of experiments performed in forced
circulation. From this value, a corresponding heat flux has been
calculated dividing the power by the bayonet outer surface area;
this heat flux (11.485 kW/m2) has been imposed as boundary
Eq. (6)
The equivalent heat transfer coefficient can be derived from the
radiative heat flux as
ℎL:M =
8 − :
Eq. (7)
Copyright © 2017 ASME
Downloaded From: on 10/25/2017 Terms of Use:
condition for the code on the external side of the heat structure
which simulates the bayonet outer tube (Heat Structure 02,
Figure 4).
Below, the electrical and thermal power transient, the
temperatures history, the absolute pressure and the pressure
drops results are reported.
is lower than the saturation value at the bayonet outlet, which is
the hottest position in the loop, the flow is liquid single-phase.
The code can predict well both the general behavior of the
experimental transient and the final steady state value. The
experimental temperature during the central part of the transient
(1000 – 10000 s) is lower than the simulated one; this can be due
to the underestimation of the total system thermal inertia
(possibly related to flanges and supports) and to the ambient air
temperature variation during the transient (that also affects the
thermal losses to the ambient). Moreover, also the positioning of
the thermocouple can affect the measurements. This behavior is
present in all the measured temperature locations. Both in the
experimental and simulated results after 270 s (4.5 min) the
temperature peaks, afterwards slightly decreases and then it
starts to increase again. This phenomenon is likely due to the
entrance of cold fluid from the cold leg after the beginning of
circulation of water in the system.
Electrical and thermal power transient
The reference decay heat removal system operates with an
imposed temperature on the external side of the bayonet heat
exchanger bundle. Since for technical reasons (toxicity and cost
of commercial high temperature diathermic oils) it was not
possible to operate in the same way, power is provided using high
temperature electric heating tapes; the electric power supply to
the facility (1.75 kW on average) is shown in Figure 6. In the
same figure the RELAP5 results concerning the thermal
incoming power in the bayonet heat exchanger are reported; the
steady state value of 1.225 kW is the one used to calculate the
heat flux boundary condition. The ratio between the two values
is 0.7 so it is possible to estimate the power losses of the heating
tapes to be 30% of the electric power in these conditions (without
thermal insulation of the bayonet).
Figure 7: Bayonet outlet temperature T2 and saturation temperature
time history
Figure 8 shows the temperature at the cold leg inlet.
Figure 6: Electric power supply to the heating tapes and thermal power
received by the bayonet in RELAP5 simulation time history
Temperature history
Water temperature comparison between the experimental
and the simulation results is presented in locations 2, 5, 7 and 8
(see Figure 3) that are significant to understand both the behavior
of the facility and physical phenomena occurring, as well as the
capability of the code to predict the physical phenomena
All the temperature measurements have a constant error of
±1°C as reported in Table 1. The error bars are not shown in the
figures since, due to the y axis scale, they would not be visible.
Figure 7 shows the bayonet heat exchanger outlet
temperature comparison and the saturation temperature
predicted by the code in the same position. Since the temperature
Figure 8: Cold leg inlet temperature T5 time history
Copyright © 2017 ASME
Downloaded From: on 10/25/2017 Terms of Use:
This temperature is very close to the results found for the
hot leg outlet (location 4 in Figure 3) since the distance between
the two points is small. The about 1°C steady state temperature
difference between the experimental and the simulated values is
acceptable; this proves the good modeling of the system, thermal
losses included, and the capability of the code to predict the wall
The cold leg outlet temperature is presented in Figure 9; the
global behavior is similar to the previous results, and the
agreement between the code predictions and the experimental
results is good also in this case. The steady state difference,
which is lower than 2.5°C, is probably due to the operating
conditions in the condenser, where air is stagnant and circulation
is limited. Circulation of air in natural convection is simulated
by the code with some difficulties and further studies should be
performed on these components, possibly by CFD codes.
not show any oscillation, because the code averages the
temperature in every volume; anyway, the predicted steady state
value is in the middle of the experimental oscillation band.
Figure 10: Bayonet heat exchanger inlet temperature T8 time history
Figure 11 shows the experimental water temperature in the
bayonet inversion chamber T1, at the bayonet outlet T2 and the
outer wall bayonet temperatures TA, TB and TC (see Figure 3).
Figure 9: Cold leg outlet temperature T7 time history
Even though the distance between measurement locations 7
and 8 (see Figure 3) is relatively short, Figure 10 presents the
results for the bayonet inlet temperature since it shows a
particular experimental behavior. With respect to all other
temperature measurements, in fact, significant oscillations
higher than 3.5°C are present in the experimental results. Since
the temperature decrease between positions 7 and 8 cannot be
just due to thermal losses (the relative distance is lower than 30
cm), this temperature behavior may be due to the presence of a
tee-junction upstream the measurement location. Since the teejunction is connected with a horizontal pipe filled with stagnant
water and recirculation is limited inside it, the temperature in the
tee-junction is lower than the temperature of the water that exits
from the condenser. Close to the tee-junction, thermal
stratification occurs and hot water enters the junction in the upper
part while relatively cold water from the junction enters from the
lower part in the loop. The effect of this local recirculation is that
when a cold liquid trickle touches the thermocouple tip there is
a temperature oscillation and the average temperature in that
position is lower. Figure 10 shows that the code predicts quite
well the average temperature behavior; the predicted values do
Figure 11: Experimental bayonet heat exchanger water temperatures T1
and T2 and outer wall temperatures TA, TB and TC time history
The minimum temperature difference between bayonet wall
and water occurs in the bottom part of the annulus riser (29°C),
as expected; the measured external temperatures increase along
the bayonet because the inner water temperature increases. It has
to be remarked that the measured external temperatures are not
exactly the bayonet outer wall temperatures, since wall
thermocouples are placed between the bayonet wall and the
heating tapes, so they are affected by the presence of the tapes
themselves, which have a higher temperature than the bayonet
wall. Moreover, as already mentioned, the heat flux supplied by
the heating tapes is not uniform but it is higher in correspondence
of every coil and lower in the bare areas (Figure 2). Also the
Copyright © 2017 ASME
Downloaded From: on 10/25/2017 Terms of Use:
external bayonet wall temperature is not uniform but increases
along the bayonet length since the outer thermal flux is imposed.
However, since the wall thickness is sufficiently high, the inner
wall temperature, which is the one that affects the water
circulation, is not affected by the non-uniformity of the heat flux,
so the measured water temperatures are in agreement with the
code predictions as already shown.
pipe 232 in Figure 4). This has been confirmed by filling and
discharging the water in the loop several times and measuring its
weight. Comparing the measured values to the simulation
results, the average error is lower than 0.3%.
Pressure drop measurements have been performed in parts
of the facility (bayonet downcomer, hot leg and cold leg) with a
relevant L/d ratio. Due to power limitations, the natural
circulation water flow rate in the facility is quite low (it will be
further discussed) and the velocity too. Therefore, also
irreversible pressure losses (both major and minor pressure
losses) are very small: their magnitude is in the order of few
Pascal. Consequently, the pressure drop in the analyzed
components is, in practice, mainly due to gravitational
contribution, and when the temperature changes over the time,
also the pressure drop modifies. As general convention, the
pressure drops have always been measured and calculated
following the flow direction.
Figure 13 shows the pressure drop between the inlet and the
exit of the downcomer pipe of bayonet, which corresponds to the
pressure drop Δp1 (see Figure 3). In this case the error of the
experimental results is ±0.05 mbar as reported in Table 1.
Absolute pressure and pressure drops
Absolute pressure is a crucial parameter to be measured in
the facility since it helps to estimate the distance to saturation
conditions, coupled with temperature measurements, and it must
be kept under control to avoid damages in the facility..
Absolute pressure is measured in two different locations in
the experimental facility. Figure 12 shows the absolute pressure
in the bayonet inversion chamber (location 1, Figure 3), which
corresponds to the highest pressure value since it is measured in
the lowest region of the facility. The error of the experimental
results is ±4 mbar as reported in Table 1. Error bars are not shown
since, due to the y axis scale, they would not be visible
Figure 13: Bayonet inlet and downcomer pressure drop ∆p1 time history
Figure 12: Absolute pressure bayonet heat exchanger inversion
chamber p1 time history
It is possible to notice that the pressure drop is negative
(pressure increase), since the flow is downwards and irreversible
pressure losses are negligible with respect to the gravitational
contribution. The code can predict well the experimental steady
state value and the global transient behavior.
As previously described, during the transient, the
experimental temperatures are lower than the simulated ones and
this is reflected in a lower density reduction rate with respect to
the code prediction; the gravitational pressure change can be
defined as
The agreement between the experimental results and the
simulations is very good; in fact, the code predicts correctly both
the pressure transient and the steady state value, which is very
close to 3.9 bar. This pressure value corresponds to a saturation
temperature of 142.7°C, so it confirms that there is liquid singlephase within the loop as expected. In fact, the initial water mass
is too high to reach saturation conditions with this temperature
level. In order to study two-phase flow natural circulation it will
be necessary to decrease the pressure through controlled liquid
bleeding and/or to insulate both the heating tapes and the loop
pipes to reduce the thermal losses and increase the temperature.
The good agreement between the experimental results and
the simulations is also due to the correct and accurate volumetric
reproduction of the facility, even in the components where air is
present (upper part of annulus 214, upper part of pipe 230 and
∆QL:( = ∆
Eq. (9)
where ρ is the average density in the considered section and ΔH
is the height difference; thus, the rate of change of the
experimental pressure difference is lower than the estimated one,
Copyright © 2017 ASME
Downloaded From: on 10/25/2017 Terms of Use:
Mass flow rate
The mass flow rate is one of the clue parameter in passive
safety system for the removal of the decay heat. Its measurement
is therefore fundamental to study the performance of the safety
system in general, and specifically the operation of the
experimental facility, as well as to compute the power transferred
from the bayonet tube to the flowing water. Unfortunately,
pretest simulations [4] showed very low values of mass flow rate,
which are hardly measurable with acceptable accuracy.
Moreover, accordingly to the code predictions, the mass flow
rate is expected to further reduce when passing from liquid
single-phase flow to two phase flow, which is how the reference
system is expected to operate. Therefore, no mass flow meter has
been installed yet in the facility and further efforts are needed to
develop a method for the mass flow rate measurement.
Figure 16 shows the RELAP5 mass flow rate prediction in
the bayonet riser, which is expected to be quite accurate since all
the other measured parameters are well predicted by the code.
and the pressure drop in the transient is more negative than the
code prediction.
Figure 14 presents the pressure drop in the hot leg, which
corresponds to the pressure drop Δp2 in Figure 3. In this case the
error of the experimental results is ±0.032 mbar as reported in
Table 1. The general behavior is very similar to the bayonet
pressure drop, even though with reversed sign, since in this case
the flow is upward and actually a pressure decrease occurs. The
code predicts the initial transient accurately, even though a slight
disagreement occurs for the final steady state value around 1
mbar that is considered to be acceptable.
Figure 14: Hot leg pressure drop ∆p2 time history
The last pressure drop measurement is performed in the cold
leg (Δp3 in Figure 3); the results are shown in Figure 15.
Figure 16: RELAP5-3D mass flow rate prediction in the bayonet heat
exchanger riser
Once the power is activated the circulation in the system
starts, then around 370 s a peak occurs, and after a slight
decrease, the mass flow rate starts to rise again up to the final
steady state value of 0.014 kg/s. This behavior is similar to what
observed for the bayonet outlet temperature T2 (Figure 7); at the
peak time there is a partial mixing of the fluid that decreases the
temperature and consequently increases the average hot leg
density, decreases the natural circulation driving force and the
natural circulation mass flow rate. The maximum of the mass
flow rate occurs almost simultaneously to the temperature
minimum occurrence at the bayonet.
Figure 15: Cold leg pressure drop ∆p3 time history
In this case the error of the experimental results is ±0.014
mbar as reported in Table 1. In the cold leg the flow is
downwards as in the downcomer of the bayonet heat exchanger
and the general behavior is very similar to that case. In fact, the
code can predict well the general transient and the steady state
value and the disagreement is due to the lower rate of density
reduction in the experiments with respect to the code predictions.
In this work passive heat removal relying on liquid singlephase natural circulation has been studied both experimentally
and using the code RELAP5-3D.
Copyright © 2017 ASME
Downloaded From: on 10/25/2017 Terms of Use:
Starting from the reference decay heat removal system
DHR2 of the lead cooled fast nuclear reactor ALFRED, an
experimental facility has been designed and built in order to have
a comparable behavior; the fundamental element of the facility
is the bayonet heat exchanger, where the heat source of the
system is located. The facility has been designed to operate in
two-phase flow regime as the reference system, but, in this first
experimental campaign, only single-phase flow has been studied
to understand the behavior of the system as a base for further
studies involving two-phase flow. In this work natural circulation
has been analyzed only in the experimental facility without any
comparison to the reference system that is still under design;
future works will be focused on the scaling criteria between the
two systems.
The facility operation has been simulated by the code
RELAP5-3D and temperatures, absolute pressures and pressure
drops have been measured to verify the predictions of the code.
Heat transfer coefficients over a vertical cylinder in natural
convection have been calculated as a function of the wall
temperature in a general case and implemented in the RELAP53D input to simulate thermal losses. For every analyzed
parameter there is a good agreement between the experimental
and predicted results; the final steady state value is always
predicted accurately, the transients are globally well reproduced
and the main differences can be related to differences in
modeling the system thermal inertia. An important remark is
that, when modeling experimental facilities with a relatively
small power, thermal losses cannot be neglected and their correct
modeling is crucial to obtain reliable results from the RELAP53D code.
Pre-test simulations have highlighted the fact that the natural
circulation mass flow rate is very low and hardly measurable and
a method for the mass flow rate measurement needs to be
The next experimental campaign foresees to decrease the
system pressure through controlled water bleeding and to
insulate the facility to reduce the thermal losses and to increase
the temperature, so that two-phase flow natural circulation will
occur. Water will be introduced into the condenser shell side as
heat sink, similarly to the reference decay heat removal system.
The RELAP5-3D input will be modified to take into account
these variations and it will be used to perform both pre-test
simulations and post-test comparisons with the experimental
results; the aim is to validate the code capabilities in natural
circulation two-phase flow, also in presence of non-condensable
gases, since this new code version showed problems when
modelling systems at low pressure and high void fractions.
We thank Marco Caramello, PhD student, for the
suggestions regarding the modeling of the facility thermal losses.
The present research has been supported by ENEA and by the
Italian Ministry of Economic Development.
[1] Alemberti, A., Frogheri, M. L., Hermsmeyer, S., Ammirabile,
L., Smirnov, V., Takahashi, M., Smith, C. F., Wu, Y., Hwang, I.
S., 2014, “Lead-cooled Fast Reactor (LFR) Risk and Safety
Assessment White Paper”, Generation IV International Forum.
[2] Grasso, G., Petrovich, C., Mattioli, D., Artioli, C., Sciora, P.,
Gugiu, D., Bandini, G., Bubelis, E., Mikityuk, K., 2014, “The
core design of ALFRED, a demonstrator for the European leadcooled reactors”, Nuclear Engineering and Design, 278, pp. 287301.
[3] Alemberti, A., Frogheri, M., Mansani, L., 2013, “The lead
fast reactor: demonstrator (ALFRED) and ELFR design”,
Presentation at International Conference on Fast Reactor and
Related Fuel Cycles: Safe Technologies and Sustainable
Scenarios, Paris.
[4] Bertani, C., De Salve, M., Caramello, M., Falcone, N.,
Bersano, A., Panella, B., 2016, “Preliminary numerical studies
of an experimental facility for heat removal in natural
circulation”, Proc. 34th UIT Heat Transfer Conference, Ferrara,
Italy, Ed. S. Piva and G. Cassano, pp. 334-343.
[5] Pitts, D.R., Sissom, L. E., 1982, Trasmissione del calore, pp.
200, Gruppo Editoriale Fabbri - Bompiani, Sonzogno, Etas.
[6] McQuillan, F. J., Culham, J. R., Yovanovich, M. M., 1984,
“Properties of Dry Air at One Atmosphere”, Microelectronics
Heat Transfer Lab, University of Waterloo, Waterloo, Ontario.
Copyright © 2017 ASME
Downloaded From: on 10/25/2017 Terms of Use:
Без категории
Размер файла
1 351 Кб
icone25, 67647
Пожаловаться на содержимое документа