th

10 International Conference on Thermal Engineering: Theory and Applications

February 26-28, 2017, Muscat, Oman

Solution of the Reactor Point Neutron Kinetic Equations with

Temperature Feedback Control Using MATLAB-Simulink Toolbox

Mai Ismail1, Dana Abulaban1, Filippo Genco1, Mohammad Alkhedher1

1
Al Hosn University, Department of Mechanical and Industrial Engineering, Abu Dhabi, UAE

Abstract
Point kinetics equations are a system of stiff nonlinear ordinary equations. The solution of these equations provides a
full understanding of neutron population’s behavior, which helps controlling the criticality of a nuclear reactor. Solving
point kinetics equations numerically has been always challenging due to the stiffness problem. In this paper, a solution for
one-group delayed neutrons with negative temperature feedback using Simulink toolbox of MATLAB is presented. The
solution is applied to the kinetics equation using initial step reactivity ρ0=0.2β. A GUI controller is built to control the
model’s input parameters. The results of the system are presented and compared with numerical methods. The variation
along time of reactivity, temperature, and neutron density is analyzed.

Keywords: Point kinetic Equation, Simulink MATLAB, Negative Temperature feedback, One-group delayed neutrons
respect to time into differential equations with respect to
reactivity. The numerical solution is based on Taylor’s
Series method. Hadad et alii (2004) [3] have developed a
1. Introduction MATLAB/Simulink model for six groups delayed neutrons
with temperature reactivity feedback and input reactivity
The study of neutron population is of great of 1.2$ and 3$. Rashid (2015) [4] reduced the point
importance in nuclear reactor kinetics. That is because it kinetic equation of six group delayed neutron equations
plays an essential role in sustaining the fission reactions into one effective delayed neutron group with and without
in a nuclear reactor and it is a priority in nuclear safety .In temperature feedback and presented them as Simulink
the start-up process of a reactor, changes in the external model. The model consists in two subsystems, prompt
source of neutrons have a significant influence on the neutron subsystem and delayed neutron subsystem at
reactor power. If control rods are withdrawn in a constant neutron precursor concentration with an initial
discontinued manner, the reactivity will increase and will step reactivity of 0.3 β. Hend et alii (2014) [5] presented a
cause a high increase in reactor power resulting in an solution for point kinetic equations with one group of
over pressurized reactor core. To study the behavior of delayed neutron using FORTRAN and MATLAB codes,
that focused on the prompt super-critical process under
neutron population, one must solve the reactor point
two cases for three initial power inputs. First case is large
kinetics equations (PK). These equations are a system of
step reactivity insertion of 1.5 β and 2 β. Second case is
stiff non-linear ordinary equations. They express the
small step reactivity insertion of ρ0= (β/4) and ρ0= (β/1.2).
time-dependence of neutron population, and the change Other reactivity shapes such as ramp and sinusoidal
of reactivity and temperature within the reactor due to a were tested as well.
change in neutron density and delayed neutron precursor In this work, we present a complete, precise, and
concentrations. A change in neutron density affects the clearer model of the point kinetics equations with one-
system leading to a “temperature-reactivity feedback”. group delayed neutrons and negative temperature
In previous works, various methods were attempted feedback using Simulink toolbox of MATLAB. A GUI
for solving the system. Hend et alii (2013) [1] presented controller is built using MATLAB script to easily change
an analytical solution for the point kinetic equation with the input parameters of the model. The results of the
one group delayed neutron in the presence of Newtonian system are discussed and benchmarked with previous
negative temperature feedback for different step works [1] [2].
reactivities. SCM (Stiffness Confinement Method) was
used to solve the system and to analyze accidents in
multiple types of reactors, e.g. thermal reactor such as 2. MATLAB-Simulink application point kinetic
HTR-M, and fast reactor like PRISM. Nahla and Zayed equation solution
(2010) [2] have introduced an analytical approximation
and numerical solution of the point nuclear reactor kinetic The point kinetics equations are stiff nonlinear
equations with average one-group of delayed neutron ordinary differential equations, which are written in the
and temperature feedback. The analytical approximation following form along with one-group delayed neutrons
is based on transforming the differential equations with equation [2]:
dn ρ (t) − β (1) dT (t ) (4)
= n(t) + λC(t) = K c n(t )
dt Λ dt

dc(t) β (2) where T(t) is the temperature of the reactor (K), T0 is the
= n(t) − λC(t) initial temperature of the reactor (K), Kc is the reciprocal
dt Λ
of the thermal capacity of the reactor (K/MW*sec), and α
-1
is the temperature coefficient of reactivity (K ).
Where n(t) is the neutron density (or power in MW), t is Initial condition of delayed neutron precursor density
the time, ρ(t) is the reactivity, β is the fraction of delayed
is introduced through the following form [2]:
neutrons, λ is the decay constant, C(t) is the density of
delayed neutron precursors (or latent power in MW), and
Λ is the generation time. ( β − ρ0 ) (5)
C (0) = n0
Negative temperature feedback is also considered Λλ
as part of the system [2]:
Equations 1 to 5 are shown in Figure (1)
ρ (t ) = ρ0 − α [T (t ) − T0 ] (3)

Fig. 1. Simulink modeling of point kinetic equation with negative temperature feedback and detailed subsystem of one group
delayed neutrons
 using Simulink toolbox of MATLAB. The results are
2.1 Simulink Model compared against other methods: SCM (Stiffness
Confinement Method) [1] and Taylor’s series method [2].
Our Simulink model provides a complete and clearer It is assumed that the input parameters are as
-1
representation of the point kinetics equations for one- following: λ=0.07741 sec , β=0.0065, Λ=0.0001 sec,
-5 -1
group delayed neutron with negative temperature K=0.05 K/(MW sec), α =5e K , n(0)=10 MW, ρ0=0.2β
feedback. It clarifies the variables and constants used in and C(0)=6717.5 MW, T(0)=300 K. In the start-up
solving the system. The main part of the system subcritical process, the external source of neutrons is
represents Equations 1, 3 and 4. Equations 3 and 4 are important and the effect of temperature feedback is
for “Negative temperature feedback.” The subsystem neglected because of reactor core’s low average
“One-Group Delayed Neutron” is a representation of temperature. During this process, neutron density is
Equations 2 and 5. The subsystem “1/s” is the initial taken to full power and expected to increase to an initial
condition of delayed neutron precursor density. Variables
maximum level due to the discontinued withdrawal of
that can be controlled through GUI are the initial step
control rods. When step reactivity is suddenly inserted
reactivity, initial condition of precursor density, initial
into the reactor at t=0, the prompt supercritical process
power and simulation time. Simulation results can be
viewed directly through scope, and indirectly using starts taking place. The power of this process increases
MATLAB script through “to Workspace” block. abruptly, and the maximum level of power is always
larger than the initial power.
2.2 Graphical User Interface GUI In Figure (4), it can be seen that reactivity begins
decreasing slowly with time due to the effect that control
Graphical user interface was built as an additional rods have on neutron population. At t>0, changes in
tool that helps more in understanding and using the reactivity act back on neutron density with negative
Simulink model. Figure (2) shows the main window of a temperature feedback slowly influencing the system.
small step reactivity GUI: it includes all the main
parameters in the Simulink model. The interface interacts
with the system so that the user can change the desired
parameters and obtain different outcomes like reactivity,
power and temperature plots. Figure (3) shows the
detailed interface of the parameters window. The
additional parameters of the Simulink model are in this
window. Also, the initial power can be modified to get
different graphical outcomes. Different scenarios in
Figure (3) can also be added to the model and the user
can move from one scenario to another in a very
convenient way.

Fig. 4. Reactivity variation along time for ρ0= 0.2β

In Figure (5), when step reactivity is inserted in the

reactor core (withdrawing control rods), temperature rises
slowly with time as a result of an increment in neutron
population and fission chain reactions.

Fig. 2. GUI for step reactivity insertion

Fig. 3. GUI parameters window

3. Simulation discussion and benchmarking

The solution of point kinetic equation is preformed Fig. 5. Temperature variation along time for ρ0= 0.2β
Table. 1. Variation of Reactivity, temperature and neutron density along time for initial step reactivity ρ0= 0.2β.

Reactivity ($) Temperature (K) Neutron Density (MW)

Time
(sec) Present Present Present
Ref [1] Ref [2] Ref [1] Ref [2] Ref [1] Ref [2]
Work Work Work
0 0.2 0.2 0.2 300 300 300 10 10 10
10 0.159 0.1589 0.158976 305.3 305.3 305.333 11.271 11.28 11.270597
20 0.1139 0.1139 0.113885 311.19 311.2 311.195 12.097 12.11 12.094137
30 0.0666 0.06659 0.06661 317.3 317.3 317.341 12.4091 12.4 12.402006
40 0.0191 0.0191 0.019115 323.5 323.5 323.515 12.2248 12.22 12.220168
50 -0.0269 -0.02687 -0.026878 329.49 329.5 329.494 11.6418 11.63 11.639050
60 -0.0701 -0.07006 -0.070057 335.1 335.1 335.107 10.778 10.78 10.777035
70 -0.1096 -0.1096 -0.109569 340.2 340.2 340.244 9.7628 9.751 9.750800
80 -0.145 -0.145 -0.14498 344.8 344.8 344.847 8.6599 8.659 8.658279
90 -0.1762 -0.1762 -0.176181 348.9 348.9 348.904 7.5733 7.573 7.572253
98.7 -0.2 -0.2 -0.2 352 352 352 6.6757 6.673 6.671694

In Figure (6), the neutron density of the delayed [2] proving that the model well responds and interprets
supercritical process increases largely reaching a correctly the physics. This model can be integrated with
maximum level. As the negative temperature feedback different scenarios or fuzzy logic tools modeling the
becomes more and more pronounced, neutron density position of control rods. A six group delayed neutrons
starts dropping with time. model has been also developed increasing the accuracy
The comparisons of the results obtained between of the solutions: the model will be presented in details in
time, reactivity, temperature, and neutron density using the extended version of the paper.
Simulink with SCM [1] and Taylor’s series [2] are
presented in Table 1. The results indicate excellent Acknowledgments
agreement with the previously published ones [1] [2].
The authors thank all the participants to the Tenth
International Conference on Thermal Engineering:
Theory and Applications. The authors would like to thank
also Giacinto Genco from KFUPM for his help and very
valuable suggestions on the topic.

References

[1] Hend M. Saad, M. Aziz, H. M. Mansour (2013).

“Analysis of Reactivity–Initiated Accident for Control
Rods Ejection”, Journal of Nuclear and Particle Physics,
45-54.

[2] A.A Nahla, E. M. E. Zayed, "Solution of the nonlinear

point nuclear reactor kinetics equations", Prog. Nucl.
Energy, P. (1-4), (2010).

[3] K.Hadad, A.Mohamadi, H.Sabet, N.Ayobian, M.Khani

Fig. 6. Neutron density variation along time for ρ0= 0.2β (2004). “Numerical Solution of Multigroups Point Kinetic
Equations by Simulink Toolbox of Matlab Software”. 2nd
International Conference on Nuclear Science and
Technology, (pp. 1-4). Iran.
5. Conclusions
Simulink model of point kinetic equation for one- [4] N. Rashid (2015). "Modeling Nuclear Processes by
group delayed neutrons with negative temperature
Simulink", in AIP Conference Proceedings, Selangor,
feedback is designed to solve the point kinetic equation.
The simulated results are plotted, tabulated, and Malaysia.
compared with previous works [1] [2] that utilized
numerical methods. The comparison of the Simulink [5] Hend M. Saad, M. Aziz, H. M. Mansour (2014).
model proposed shows excellent agreement with the “Analysis of Reactivity accident for control rods
results of previous works [1] withdrawal at the thermal reactor”. International Journal
of Recent advances in Physics (IJRAP), 21-40.