NUKLEONIKA 2015;60(3):537544

doi: 10.1515/nuka-2015-0110 ORIGINAL PAPER

Thermal-hydraulic calculations Maciej Skrzypek,

Rafał Laskowski
for a fuel assembly
in a European Pressurized Reactor
using the RELAP5 code

Abstract. The main object of interest was a typical fuel assembly, which constitutes a core of the nuclear reactor.
The aim of the paper is to describe the phenomena and calculate thermal-hydraulic characteristic parameters
in the fuel assembly for a European Pressurized Reactor (EPR). To perform thermal-hydraulic calculations, the
RELAP5 code was used. This code allows to simulate steady and transient states for reactor applications. It is
also an appropriate calculation tool in the event of a loss-of-coolant accident in light water reactors. The fuel
assembly model with nodalization in the RELAP5 (Reactor Excursion and Leak Analysis Program) code was
presented. The calculations of two steady states for the fuel assembly were performed: the nominal steady-state
conditions and the coolant flow rate decreased to 60% of the nominal EPR flow rate. The calculation for one
transient state for a linearly decreasing flow rate of coolant was simulated until a new level was stabilized and
SCRAM occurred. To check the correctness of the obtained results, the authors compared them against the reac-
tor technical documentation available in the bibliography. The obtained results concerning steady states nearly
match the design data. The hypothetical transient showed the importance of the need for correct cooling in the
reactor during occurrences exceeding normal operation. The performed analysis indicated consequences of
the coolant flow rate limitations during the reactor operation.

Key words: fuel assembly • pressurized water reactor (PWR) • safety analysis • RELAP5

European Pressurized Reactor

The European Pressurized Reactor (EPR) is an light

water reactor in which light water H2O is used as
moderator (for slowing-down neutrons) and core
coolant (for heat removal). The EPR was designed
by a French company AREVA on the basis of experi-
ence gained during many years of operation of N4
and KONVOI reactors. Electric power generation in
a nuclear power plant with the EPR is similar to that
M. Skrzypek
of a conventional power plant with ranking cycle. A
Institute of Heat Engineering,
significant difference between these two systems is
Warsaw University of Technology,
visible during the system shutdown. After the reac-
21/25 Nowowiejska Str., 00-665 Warsaw, Poland
tor shutdown in a nuclear power plant, the so-called
and National Center for Nuclear Research (NCBJ),
decay heat is generated, as a result of decay of fission
7 Andrzeja Soltana Str., 05-400 Otwock/Swierk,
products and actinides. Although it decreases rela-
Tel.: +48 22 273 1430,
tively fast, as shown in Fig. 1, it is a major concern
E-mail: maciej.skrzypek@itc.pw.edu.pl
for the power plant operators. The uncontrolled de-
R. Laskowski cay heat and disability of the core structures cooling
Institute of Heat Engineering, are the main reasons for core damages and severe
Warsaw University of Technology, accidents. The problems created by nonremoval
21/25 Nowowiejska Str., 00-665 Warsaw, Poland of decay heat were visible during the events in the
Fukushima Daichii Nuclear Power Plant in 2011,
Received: 24 September 2014 when the damage to the reactor cores and the release
Accepted: 20 May 2015 of the radioactivity necessitated evacuation. For
538 M. Skrzypek, R. Laskowski

different power levels, due to their location in the

core and enrichment.
Safety analyses concerning the EPR fuel as-
sembly are allowing to investigate safety margins
during normal operation and operation with lower
coolant flow rate or a hypothetical failure of pumps,
leading to decreased coolant flow rate. Calculations
were performed using RELAP5, a thermal-hydraulic
calculation code [3].

Fuel assembly

The EPR fuel assembly is built as a square grid made

of 17 × 17 rods, including 265 fuel rods and 24 guide
rods for control rods or measuring apparatus. As
Fig. 1. Decay heat in an EPR [2]. fuel, the EPR uses uranium dioxide UO2 enriched to
2.25–3.25 wt% of the entire rod. Some fuel rods also
safety reasons, mentioned heat must be removed [1]. contain gadolinium oxide Gd2O3, which will ‘burn
After the reactor shutdown, that is, after neutron out’, that is, absorb neutrons to enable more stable op-
absorber rods are inserted, over 7% of the reactor eration of the reactor during the first fuel loading. The
nominal power (decay heat) is still generated, which gadolinium content varies from 2 to 8% depending on
in the case of the EPR equals about 315 MW [2]. the location of the fuel assembly in the reactor core.
Because the shutdown reactor and spent fuel pools Apart from fuel rods and guide rods, connected
need to be cooled constantly, all necessary efforts with spacer grids, the fuel assembly also contains
shall be made to provide cooling water even if elec- inlet and outlet connections and springs for fixing
tric power for driving pumps is no longer supplied. the component in the core (Fig. 2).
The EPR belongs to a group of pressurized water The fuel rod consists of fuel pellets kept in a
reactors (PWR), which have two circuits: primary thin-walled cladding made of zirconium alloy. An
and secondary ones. Its distinctive features are innovative M5 alloy, which apart from zirconium
modern detection devices, safety systems and special also consists of niobium (1%), oxygen, and iron,
core catcher, that is designed to keep the molten was used by the reactor manufacturer as the clad-
corium outside of the reactor pressure vessel, but ding material. Since the alloy contains no tin as an
in control over its temperature and high decrease alloy-forming element, it is highly corrosion proof,
of the radioisotopes (in the form of volatile species) as M5 features lower hydrogen production resulting
releases to the containment atmosphere. The reactor from zirconium oxidation. A cross section of the fuel
core itself consists of 241 fuel assemblies, being at rod is pictured in Fig. 3.
The fuel rods and guide rods are connected
with each other in the form of a 17 × 17 matrix by
10 spacer grids, part of which is shown in Fig. 4.
The spacer grids also have another function: as
water flows between them, it is mixed better and
put into whirling motion, which helps remove the
heat more efficiently.
While coolant flows through the fuel assembly,
its pressure decreases with subsequential increase
in temperature. Pressure drop results from the resis-
tance of the coolant flow through the fuel assembly.
An average pressure drop in the EPR fuel assembly is
about 0.188 MPa. As the fission energy is absorbed,
the temperature of the coolant flowing through the
fuel assembly increases by about 36 K. The cooling
water flow rate through one fuel assembly is about
96.097 kg/s. At standard rating conditions, about
18.672 MW of power is generated, on an average,
Fig. 2. Fuel assembly [4, 5]. in one fuel assembly [4].

Fig. 3. Fuel rod [4].

Thermal-hydraulic calculations for a fuel assembly in a European Pressurized Reactor... 539

Fig. 4. Part of spacer grids [4].

RELAP5 code
Fig. 6. Nodalization and the end view of a fuel assembly
RELAP5 is a code designed to perform thermal- [4].
-hydraulic calculations concerning light water reac-
tors (LWR), that is, for fluids such as water, steam is a commercial code that was verified and validated
and water mixture, noncondensible gases, and non- for nuclear applications in the past 30 years.
volatile matter (boron). The code includes modules The calculations are performed separately for
dedicated to reactors, particularly a point kinetics each control volume, as the tool solves equations
model, pumps (including a jet pump typical of boil- relating to mass, momentum, and energy balances
ing water reactors), valves, pipes, heat structures, for each phase of the fluid. When creating control
turbine, separator, water accumulator, and logical volumes, one has to adhere to certain guidelines
elements for system control. The code provides ap- and rely on one’s experience in modeling complex
plications typical of LWRs, simulating small coolant thermal-hydraulic systems. In defining the sizes
loss, anticipated transients without SCRAM, power of the control volumes, the geometry complexity and
outage, and loss of flow. The code was developed at the rate of changes in basic parameters within the ge-
the U.S. Idaho National Laboratory [3]. ometry need to be taken into account. Nodalization
(a division into control volumes) of a fuel assembly
as a whole is demonstrated in Fig. 6.
Fuel assembly nodalization Once created, the fuel assembly nodalization con-
sists of a lower and upper source, a time-dependent
In order to input geometrical data, the fuel as- volume, where the pressure, temperature, and void
sembly geometry has to be discretized into control fraction are declared. Other components required to
volumes; the connected control volumes constitute model the fuel assembly is a time-dependent junc-
a calculation model. A sample division into control tion, branch, and pipe. To model the process of heat
volumes for a pipe can be found in Fig. 5. RELAP5, exchange and generation, one has to introduce heat
being a system code, can make the impression that structures. To make the tool differentiate between
the number of components and control volumes fuel rods and control rods, two heat structures, both
is small (when compared to techniques used in marked in red in Fig. 6, were added to the pipe;
computational fluid dynamics, or CFD software); the red-filled area on the left-hand side means that
nevertheless, such an approximation in calculations the given volume is a heating component (an active
concerning large components is adequate, and the structure).
calculations are made in relatively short time. The
time in the case of the safety analysis is crucial,
because during this kind of process, large amount Results
of calculation is needed. Nowadays, safety analyses
done with the use of the system codes are beginning Steady-state parameters
to be always done with the uncertainty and sensitiv-
ity study. Apart from the calculations of the accident In order to obtain the results for the steady state of
scenarios, results are evaluated in terms of safety the fuel assembly in the EPR operating at nominal
margins by their susceptibility to various parameter power, first, one has to check whether and when
changes. That is why the RELAP5, with its low com- the steady state is reached in the model. To this
putational time, is a great tool for performance of end, a number of variables are examined: tempera-
numerous calculations. In addition, RELAP5 code ture within fuel at half the height of the assembly,
temperature of the coolant at the outlet of some
rods and guide rods, and the pressure of the coolant
flowing out of the assembly. If these parameters vary
by less than 1%, the steady state is assumed to have
Fig. 5. Pipe discretization represented by control volumes. been reached. Table 1 lists steady-state parameters
540 M. Skrzypek, R. Laskowski

Table 1. Design and calculated steady-state parameters

Item Parameter Simulation result Design data Unit

1 Pressure at the fuel assembly inlet (lower connection) 15.6875 15.688 MPa
2 Pressure drop between the lower and upper connections 186.926 188.0 kPa
3 Overall water mass in the fuel assembly 76.165 75.0 kg
4 Height where steam appeared – – m
5 Rate of the coolant flow through the fuel assembly 96.097 92.26 kg/s
Temperature of the liquid phase of the coolant at the
6 601.84 604.85 K
upper connection of the fuel assembly
Mean flux density of the heat given off by the fuel
7 562.40 547.0 kW/m2
cladding
8 Mean velocity of the flow along the fuel assembly 5.72 5.0 m/s
Temperature in the middle of the fuel at half of the
9 1172.8 n/a K
height of the assembly

Fig. 9. Power generated in the fuel assembly.

height is shown in Figs. 7–10. Figure 7 illustrates

the change in temperature of the coolant along the
fuel assembly, in Kelvins, and the void fraction on
the right-hand side.
Figure 7 clearly shows that the flowing coolant
warms up uniformly as it passes and cools down the
Fig. 7. Temperature and void fraction of coolant. fuel rods. The fuel assembly power and flow rate
were chosen by designers so that no two-phase flow
occurs during operation at nominal power; this is
evident on right-hand side of Fig. 7, where the void
fraction is zero.
Figure 8 depicts a temperature distribution
within the fuel structure; 1 is the middle part of the
fuel, while 10 is the outer wall of the fuel cladding.
Lowest temperatures (items 8–10) represent the
temperature in the cladding, 7 is the helium gap
in the fuel rod, and the rest are the temperature of the
fuel pellet itself. Owing to the relatively low thermal
conductivity of uranium dioxide (~3.5 W/mK), the
Fig. 8. Axial temperature distribution in the fuel rod.
fuel is characterized by a large temperature gradient
obtained from the EPR fuel assembly model and the of about 550 K. However, the maximum value of
data provided by the manufacturer. 1172.8 K is very different from UO2 melting point
When the results are compared with data provid- of 2820 K [4]. The inner (item 8) and outer (item
ed by the manufacturer, they seem highly similar. The 10) temperatures of the fuel cladding are close to
most important thermodynamic parameters of the each other, since M5 is a good thermal conductor.
coolant, that is, temperature and pressure, are nearly Figure 9 illustrates the power generated in the
equal to the values found in the technical specifica- fuel assembly, broken down into characteristic com-
tions (Table 1). The right temperature obtained at ponents. The steady state occurs after 10 s, when
the assembly outlet means that the power generated the changes in the parameters in the current step
by the fuel rods and the heat transfer surface area are less than 1% over the previous step.
are properly calculated. The correctness of the pres- The black line marks the overall power gener-
sure drop calculations is proved by the pressure at ated in the fuel assembly; the red line indicates the
the assembly outlet. As for the assembly, the inlet power from fission reactions; and the green line is
pressure was declared, while the outlet one was cal- the power generated as a result of decay of fission
culated based on the right selection of roughness of products and actinides.
the assembly materials and on the local losses in the Despite the nominal power of the fuel assembly
connections and spacer grids. A detailed analysis of (18.677 MW), the flow is single-phased, and there
the steady-state parameters along the fuel assembly is no coolant vaporization. This is a consequence
Thermal-hydraulic calculations for a fuel assembly in a European Pressurized Reactor... 541

Fig. 11. Fluid temperature in the fuel assembly compared

Fig. 10. Pressure drop in each volume in the fuel assem- to the nominal parameters at 100% mass flow rate.
bly (blue bars, left axis; red bars, right axis, only for the
active fuel part).
of a very high pressure which at the fuel assembly
inlet amounts to 15.688 MPa. However, the pressure
decreases along the assembly by values indicated in
Fig. 10. The largest pressure drop occurs in the con-
nections (blue bars at the far right and left, 11 and
43 kPa, respectively) and results from the change in
the reduction of the flow surface area. The overall
pressure drop across the fuel assembly is 188 kPa.
Fig. 12. Void fraction.
Owing to the large difference between the first and
last control volumes and those in the middle, a sec-
ond axis was added to facilitate reading the pressure
drops in the active part of the fuel (volumes 2–20, red
bars). Pressure drops in the middle part correspond
with 5–7.5 kPa (right axis). Differences between
neighbor bars come from location of spacer grids
(volumes 4 and 6–18), which introduce additional
pressure drops.

Fig. 13. Power generated in the fuel assembly.

Change in the flow rate of coolant down to 60%
of the nominal value tion temperature; from the 16th volume on, steam
appeared in the flow (Fig. 12).
In this case, the analysis concerned a situation where The amount of void is so small, however, that
the coolant flow rate is immediately decreased to the heat is still properly removed from the fuel rods,
60% of the nominal value without SCRAM (reactor and damaging the fuel cladding and the fuel itself
trip). Such a situation is very unlikely, but it should is impossible.
be analyzed with respect to safety. The consequences As the moderator temperature increased, its
of such scenario for the fuel assembly parameters density drops and neutrons are slowed down less
should be investigated. The general understanding efficiently. This effect is defined by a moderator
of the phenomena present during this scenario can temperature coefficient, and with decreasing density,
be explained by the sum of the two neutronic effects it leads to negative reactivity. The drop in the flow
– the Doppler effect and the negative void fraction rate of the coolant leads to the increase in the fuel
coefficient for the EPR assembly. The effects influ- temperature; this, as a consequence of the Doppler
ence the safety-related parameters such as cladding effect, also results in negative reactivity and reduces
temperature, which is discussed in this section. To immediately the power generated in the assembly
facilitate the observation of parameter changes, the during the fission reaction (Fig. 13).
calculations were performed after a 100-second-long As in Fig. 9, the black line marks the overall
steady-state period; then, within 1 s, the coolant flow power generated in the fuel assembly; the red line
rate was decreased. indicates the power resulting from fission reactions;
The drop in the coolant flow rate led to a rise in and the green line is the power generated as a result
temperature at the fuel assembly outlet and the satura- of decay of fission products and actinides.
tion temperature was reached (Fig. 11). Line numbers The way in which the pressure changes in such a
01–20 were assigned according to the nodalization scenario should also be examined. Pressure drops in
(01, inlet; 20, outlet of the fuel assembly). each control volume along the assembly are shown
We can see that the coolant temperature rose in Fig. 14.
faster than during operation at nominal parameters It can be seen that the bars on the right-hand
and is nonuniform. The nominal outlet temperature side are clearly longer, which is associated with the
was already reached in the ninth control volume appearance of steam. It has greater velocity, which
(component number 103090000). In the five fol- leads to increased pressure drop that can be deter-
lowing volumes, heating occurred up to the satura- mined from the relation [6]
542 M. Skrzypek, R. Laskowski

Fig. 14. Pressure drop along the fuel elements.

Fig. 15. Coolant fluid temperature.

2 G2 L L
(1) p  Cf ,lo  lo2 dz  g sin    v  (1  )l 
Dh l 0 0

L d  x2 (1  x)2   N 2  G2
G 2     dz    lo,d,isi 
0 dz  v (1  )l   i 1  2l
The first, second, and third terms in the relation
(1) are the pressure drops resulting from friction,
gravity, and steam acceleration, respectively (a non- Fig. 16. Coolant gas temperature.
homogeneous model); the last one represents local
losses. The symbol 2lo is the loss factor regarding and automation system would use stand-by pumps
the two-phase flow and is greater than one. to supply the missing coolant, or the SCRAM sig-
nal would be initiated earlier. Despite many safety
measures provided in the nuclear reactor, running
The transient state: the drop in the coolant flow such a scenario is required to examine how the core
rate and the SCRAM would react if enough coolant is not supplied. Char-
acteristic parameters concerning the fuel assembly
The transient state is described by the following during a transient are listed in Table 2.
scenario. The calculations start at the steady state During a transient, physical quantities change
at nominal conditions of the reactor operation in time. To help analyze the changes in detail, they
(100 s). Then, the rate of flow of the coolant through were presented in diagrams for certain characteristic
the fuel assembly starts to decrease linearly by time points and periods. For us, the characteristic
0.5 kg/s for a period of 180 s. After this period, the time points and periods are the following: 171.5 s,
flow rate stabilizes at 6.08 kg/s, which is about 6.26% steam appears; 171.5–280 s, steam continues to ap-
of the nominal flow rate. From the beginning, the fuel pear and the power drops; 280–400 s, cooling with
assembly is heated by the nominal power but changes steam and water mixture; 400–450 s, restoring wa-
in moderator density and fuel temperature lead to ter cooling. Figure 15 shows the temperature of the
inherent power change. This results from reactivity liquid phase. As the flow rate drops, the temperature
effects related to the aforementioned safety factors rises until it reaches the value (about 618 K) match-
(particularly the higher temperature and lower den- ing the saturation pressure in the channel. This value
sity of the moderator, meaning poorer moderation is reached first at the end of the heating part. Then,
and larger proportion of fast neutrons that do not the value is reached in the preceding volumes. Dur-
contribute to fission reactions, and therefore, to the ing calculations, a temperature that is lower than the
power drop). Then, within 400 s, the SCRAM (reac- value matching the saturation pressure remains in
tor emergency shutdown) signal is given, forcing the the ninth volume. The gaseous phase temperature
fission reaction to stop by inserting safety rods (the (Fig. 16) decreases as the pressure drops. The
negative reactivity), and the reactor power decreases. amount of the heat supplied does not lead to super-
However, the flow rate of the decay heat from the heating the steam and drying the fuel cladding, which
decay of fission products and actinides remains would result in a sudden rise in the temperature of
(Fig. 1). Such a state in the nuclear power plant gas and, eventually, of the fuel cladding.
could result from the decrease in the rotational speed From the 171st second, the coolant flow becomes
of a primary-circuit pump, followed by the insertion two-phased at the end of the fuel active part. Over
of the control rods initiated by the SCRAM signal. time, the two-phase flow propagates toward the first
In the EPR, such a drop in the flow rate would be part of the assembly. The changes in the void fraction
immediately signaled to an operator and control are clearly depicted in Fig. 17. The lower the flow
rate (100–280 s), the larger is the void fraction in
Table 2. Characteristic parameters during a transient
Item Parameter Result Time [s]
1 Peak fuel temperature 1167.4 K 223.0
2 Peak fuel cladding temperature 626.52 K 227.0
3 Minimum coolant density 250.1 kg/m3 281.0
4 Coolant mass flow rate when steam appears at the end of the active part 60.68 kg/s 171.5
Thermal-hydraulic calculations for a fuel assembly in a European Pressurized Reactor... 543

Fig. 17. Void fraction. Fig. 21. Power generated in the fuel assembly.

Fig. 18. Fuel and cladding temperature. Fig. 22. Hydrogen production.

moderator temperature coefficient, which with de-

creasing moderator density (Fig. 19) makes reactivity
to attain negative values (the black line in Fig. 20).
The increase in temperature of the cladding material
did not exceed its melting point at 2100 K, but the
increased rate of cladding oxidation was reached.
This effect is presented in Fig. 22.
Owing to the poorer moderation, the reactor
power (Fig. 21) and the fuel temperature dropped.
Fig. 19. Moderator density. The fuel temperature drop leads to positive reactivity
(the red line), but after summing up the temperature
coefficients of the moderator and fuel, the reactiv-
ity is negative (0–400 s, the green line). Then, by
inserting the safety rods into the assembly, the initi-
ated SCRAM signal provides the negative reactivity
(410 s, the green line in Fig. 20).
It should be noted that in the calculations, the
RELAP5 code takes no account of the heat generated
during the fuel cladding oxidation, which is why the
actual value is greater. The oxidation of the zirco-
Fig. 20. Reactivity feedbacks. nium fuel cladding is accompanied by the production
of hydrogen, which in certain concentrations is a
the control volumes. Initiating the SCRAM (400 s) highly explosive substance. The fuel cladding oxida-
stops the fission reaction, and the power in the as- tion occurs according to the following relation [7]:
sembly is lowered more than 10 times, which results
in the appearance of water along the whole height (2) Zr + H2O = 2H2 + ZrO2 + 6500 [kJ/kg]
of the assembly.
The melting points of the materials of which the By including a specific module in the RELAP5
fuel assembly is made are the following: the fuel pel- code, we can calculate the amount of the hydrogen
let (UO2), 2820 K; the fuel cladding (the M5 alloy), produced. The integrated hydrogen production for
1450 K; the connections (stainless steel), 1454 K. To a single assembly is shown in Fig. 22. The black line
find out whether any of these points were exceeded represents the hydrogen production during opera-
in the material, we should analyze the temperature tion at nominal parameters, while the red line shows
diagram concerning the fuel element (Fig. 18). Ac- the production in the scenario under consideration.
cording to the legend in the diagram, the parameters It is evident that only about 5 × 10−9 kg H2 was
ending with one represent temperatures in the middle produced, and the relating amount of heat (0.33 J)
of the fuel in the 2nd, 5th, 10th, 15th, and 19th can be omitted. If the fuel claddings were exposed,
control volumes. Temperatures of the fuel cladding the oxidation would be much more intense and the
were assigned to parameters ending with zero. As amount of the heat generated should be included in
can be seen, the temperature of the fuel decreased, the balance, as this would result in the rise of the
while that of the fuel cladding increased by more fuel cladding temperature.
than 10 K. This happens due to the feedback of the
544 M. Skrzypek, R. Laskowski

Summary The system RELAP5 code allows to investigate

thermal-hydraulic phenomena in detail, but for
A fuel assembly model for the EPR was created using analyzing severe accidents, codes dedicated to such
the RELAP5 code. In the model, the fuel assembly phenomena should be used. The RELAP/SCADAP
was nodalized (divided into control volumes). Based extension would enable the tool also to calculate
on technical documentation, the fuel assembly the phenomena associated with processes such as
geometry, local pressure drops, and material quali- fuel cladding oxidation or hydrogen production.
ties were assumed. More professional severe accident codes, such as
Two steady-state simulations were performed: the ASTEC (Accident Source Term Evaluation Code) and
first one at nominal parameters, in order to verify MELCOR (Methods for Estimation of Leakages and
the correctness of the steady-state model, and the Consequences of Releases), also make it possible to
second one at the coolant flow rate decreased to consider the core movement, as well as the behavior
60%. The data obtained from the two steady-state of aerosols and fission products in the containment,
simulations were close to those provided by the and to evaluate the amount of their release.
manufacturer in the technical documentation.
An analysis was performed for one transient.
However, taking into account the number of safety References
systems installed in the EPR, the transient under
consideration can be described as hypothetical, for 1. Kiełkiewicz, M. (1987). Jądrowe reaktory energety-
it was assumed that no safety systems were in place czne. Warsaw: Wydawnictwo Naukowo-Techniczne.
to ensure the coolant flow and that the SCRAM 2. Pairot, F. (2011). Nuclear design. The pre-construc-
signal was delayed. Nevertheless, such an analysis tion safety report (PCSR) (Chapter 4.3).
3. Scientech, Inc. (1998). RELAP5/MOD3 code
proves the importance of these systems, and the
manual. Volume I: code structure, system models
safety of nuclear reactors, which, in the event of the and solution methods. Rockville, Maryland, Idaho
drop in the coolant flow rate inherently decreases Falls, Idaho.
their power. Safety margins, which are broad dur- 4. Framatome ANP, Inc. (2005). EPR design descrip-
ing normal operation, were also within appropriate tion. Lynchburg.
limits during the transient in question. 5. U.S. NRC. (2013). EPR final safety report (Chapter
The results are very satisfactory and form an 4.3 Nuclear design).
excellent basis for carrying further studies on the 6. Anglart, H. (2010). Thermal-hydraulics in nuclear
fuel assembly behavior during the reactor operation. systems. Stockholm: Kungliga Tekniska Högskolan.
Transients relating to the loss-of-coolant accidents 7. Cathcart, J. V. (1977). Reaction rate studies, IV, Zir-
conium metal-water oxidation kinetics. Oak Ridge
(LOCA) and the loss-of-offsite power (LOOP) can
National Laboratory. (ORNL/NUREG-17).
be particularly interesting.