Computational Modelling (PHY3709) Practical Assignments

A. E. Botha (module leader)

September 27, 2015

1 Introduction
Welcome to the practical component of the Python computational modelling module. During this week
you will each be required to complete a programming project, the results of which you will describe in a
short report. This year I have decided to pair you up into teams of two. Each team will be working on
slightly di¤erent aspects of essentially the same project (research area). The project itself is centred in
the idea of chimera states [1]. (Also see Refs. [2, 3].) Chimera states are a particular type of solution
one can obtain for systems of identical nonlinear oscillators that are coupled together, usually nonlocally.
The chimera state is basically a solution in which one group of oscillators are oscillating coherently, while
the other group oscillates incoherently. I will explain a bit about this in my introductory/welcoming
presentation.
By the way, all the references cited above, as well as a few more references related to chimera states,
will be supplied to you in electronic format. Note that, for the purposes of you report, and to make this
simpler, you should follow the referencing style that I am using here exactly. The references are sorted in
alphabetical order, by the last names of the …rst authors, and the numbering refers to the order in which
they are cited. Your report should be typeset in MSWord, LaTex, or any similar typesetting program, and
it should contain more or less the same section headings, as those used here; i.e. Introduction, Theory,
Numerical Implementation, etc.

2 Theory
The background theory for your project is mostly contained in Refs. [1, 2, 3, 4]; however, you may
supplement this list of references if you …nd any other relevant articles among those reference I have
supplied (or from anywhere else). I will also be happy to assist you in explaining any details that may
not be clear from the references I have supplied. So please ask me if you don’t understand something you
read.
Your main purpose in this project is to use Python to solve systems of coupled nonlinear ordinary
di¤erential equations: for example, Eq. (4) of Ref. [2]
Z
@
(x; t) = ! G (x x0 ) sin ( (x; t) (x0 ; t) + ) dx0 , ((4))
@t
or Eq. (1) of Ref.[4]
k+R
_ k (t) = ! 1 X
sin ( k (t) j (t) + ) . ((1))
2R
j=k R

The last two equations both have chimera state solutions, under the right conditions, as you will read
in the references.

1
3 Numerical Implementation
The basic technique for solving an equation such as (4) is to express the integral as a quadrature formula,
and then to solve the resulting set of …rst order, nonlinear, di¤erential equations. Although there are a
wide variety of quadrature rules available for evaluating such integrals, they all boil down to the following,
namely, that the integral can be expressed as a weighted sum of the integrand, evaluated on some grid.
For example, the well known Simpson’s 3=8 rule can be written as

Zb N
X
f (x)dx = wi f (xi )
a i=1

where the weights wi are given by 3h=8, 9h=8, 9h=8, 6h=8, 9h=8, 9h=8, 6h=8, . . . , 9h=8, 3h=8. In this
case h is the separation between the successive xi , given by h = (b a) = (3n), where 3n + 1 = N is the
number of oscillators and n = 1; 2; 3; : : :. So, for example, for the 256 oscillators in Fig. 1, of Ref. [3], we
would have n = 85. Incidentally, note that the caption of Fig. 1, of Ref. [3], is incorrect. It should say
= 1:457, as is written in the main text.
Thus, by using quadrature one can obtain a system of di¤erential equations, from Eq. (4), having the
following form
N
X
@
(x1 ; t) = ! wj f (x1 xj )
@t j=1
N
X
@
(x2 ; t) = ! wj f (x2 xj )
@t j=1
..
.
N
X
@
(xN ; t) = ! wj f (xN xj )
@t j=1

and in which,
f (xi xj ) = G (xi xj ) sin ( (xi ; t) (xj ; t) + ) .
The next step is to solve the above system, by integrating with respect to time, using a suitable integration
scheme. in this project we will make use of a …fth order Runga-Kutta integration scheme. To save you
some time, I have provided a Python implementation of the method on the following page. You also have
this script electronically, so there is no need to retype it.

2
from scipy import *
def f(t,x,a):
"""
Function to be integrated. Input:
t - initial time
x - vector of initial conditions at time t
v - vector of parameters, in this case (a, b, c)
"""
xp = zeros(len(x),’d’)
xp[0] = -x[1] - x[2]
xp[1] = x[0] + a[0]*x[1]
xp[2] = a[1] + x[2]*x[0] - a[2]*x[2]
return xp
def integrate(t,x,func,h,v):
"""
5th-order Runge-Kutta integration scheme.
Input:
t - initial time
x - vector of initial conditions at initial time t
h - time step
v - parameters to be optimized (in this case the period T)
p - additional parameters that get passed to the function f
Return:
xp - vector of total time derivatives at time t + h
"""
k1=h*func(t,x,v)
k2=h*func(t+0.5*h,x+0.5*k1,v)
k3=h*func(t+0.5*h,x+(3.0*k1+k2)/16.0,v)
k4=h*func(t+h,x+0.5*k3,v)
k5=h*func(t+h,x+(-3.0*k2+6.0*k3+9.0*k4)/16.0,v)
k6=h*func(t+h,x+(k1+4.0*k2+6.0*k3-12.0*k4+8.0*k5)/7.0,v)
xp = x + (7.0*k1+32.0*k3+12.0*k4+32.0*k5+7.0*k6)/90.0
return xp
def main():
fout = open(’output.dat’,’w’)
v0 = array([0.2,0.2,5.0]) # parameters
x0 = array([-10.0,0.0,10.0]) # initial condition
h = 1.0/128.0 # time step
t = 0
i = 0
while t < 10.0:
t = i*h
u0 = (t, x0[0], x0[1], x0[2])
print >> fout, (’%12.8f %20.16f %20.16f %20.16f’ % u0)
print (’%12.8f %10.6f %10.6f %10.6f’ % u0)
x0 = integrate(t,x0,f,h,v0)
i = i+1
fout.close()
print ’done!’
main()

3
4 Results
By reading the references, and using the above information, you should be able to solve either of the
given equations. Each team will however concentrate on di¤erent aspects:

Team 1 will aim to reproduce the original result by Kuramoto and Battogtokh [2], as shown in
Fig. 2 of the paper.
Team 2 will aim to reproduce the result by Abrams and Strogatz [1], shown in Fig. 1(a) of their
paper.

Team 3 will aim to reproduce the result shown in Fig. 1 of the paper by Abrams and Strogatz [3].
Team 4 will aim to reproduce the result shown in Fig. 6 of the paper by Wolfrum and Omel’chenko[4]
Team 5 will aim to reproduce the result shown in Fig. 6 of the paper by Laing[5]
Team 6 will aim to reproduce the result shown in Fig. 1 of the paper by Omel’chenko et al. [6]

Since chimera states are sensitive to the initial conditions you may …nd that your simulations do not
yield precisely the same result as that shown in the …gures that are mentioned above; however, although
the exact positions of the oscillators may vary, the basic two regions shown in the …gures should be the
same, if you have not made any mistakes in your calculations.
Figure 1 shows the chimera state reproduced for the case assigned to Team 2. As a general rule, the

Figure 1: The chimera state corresponding to the equations discussed in connection with Fig 1(a) of the
paper by Abrams and Strogatz [1], after 400 000 units of simulation time.

labels and any text in your …gure should be no smaller (and not too much larger) than the font used
in the main text of the document. The …gure itself should also be large enough to illustrate what it is
intended to illustrate. For convenience, the data as well as the plotting scrip used to generate this …gure
are supplied to you in electronic format. Incidentally, the Latex …le for this report is also supplied!

4
5 Conclusions
Chimera states are a relatively new and very interesting topic from the point of view doing further
research. Thus, after completing this project, should anyone be interested to continue research in this
direction (perhaps for an honours or masters project in a year or two), please contact let me.
I will of course be available (most of the time) to guide you through the project and to answer all the
questions that you may have. By Thursday and Friday you should be starting to write your individual
reports, which will be strictly due before you leave on Friday (unless there are unforeseen circumstances
that prevent you from completing the report before then, in which case I may allow a few extra days).
Together with your report you should also provide me with electronic copies of the scripts that you wrote
in connection with the project. These will be tested and form part of your portfolio (report plus scripts)
mark.
I hope you all have a very pleasant week at Unisa and I thank you in advance for your successful
participation in the module.

Appendix A: Additional references

Some additional references are supplied for your convenience. The references are [7, 8, 9, 10, 11, 12, 13,
14, 15, 16, 17]. Note you do not need to read all these references, they are simply included here to supply
more background information, which you may or may not need.

Appendix B: Warm-up exercise

Use the supplied script to integrate the given function for 50 time units and plot the resulting trajectory
on a graph of z (t) versus x (t), i.e. a projection onto the zx-plane. The graph should have the appropriate
labels on the axes. Print out your graph and hand it in with your student number and the value of the
parameter c (also unique to each student) written on the top right-hand corner. Time limit: 2 hours.

References
[1] Daniel M. Abrams and Steven H. Strogatz, “Chimera states for coupled oscillators,”Phys. Rev. Lett.
93, 174102 (2004)
[2] Y. Kuramoto and D. Battogtokh, “Coexistence of coherence and incoherence in nonlocally coupled
phase oscillators,” Nonlinear Phenom. Complex Syst. 5, 380 (2002)

[3] Daniel M. Abrams and Steven H. Strogatz, “Chimera states in a ring of nonlocally coupled oscilla-
tors,” Int. J. Bifurcation Chaos 16, 21 (2006)
[4] Matthias Wolfrum and Oleh E. Omel’chenko, “Chimera states are chaotic transients,”Phys. Rev. E
84, 015201(R) (2011)

[5] Carlo R. Laing, “The dynamics of chimera states in heterogeneous Kuramoto networks,”Physica D
238, 1569 (2009)
[6] Oleh E. Omel’chenko, Matthias Wolfrum, and Yuri L. Maistrenko, “Chimera states as chaotic spa-
tiotemporal patterns,” Phys. Rev. E 81, 065201(R) (2010)

[7] Daniel M. Abrams, Rennie Mirollo, Steven H. Strogatz, and Daniel A. Wiley, “Solvable model for
chimera states of coupled oscillators,” Phys. Rev. Lett. 101, 084103 (2008)

5
 [8] Juan A. Acebrón, L. L. Bonilla, Conrad J. Pérez Vicente, Félix Ritort, and Renato Spigler, “The
Kuramoto model: A simple paradigm for synchronization phenomena,” Rev. Mod. Phys. 77, 137
(2005)

[9] Tomasz Kapitaniak, Patrycja Kuzma, Jerzy Wojewoda, Krzysztof Czolczynski, and Yuri Maistrenko,
“Imperfect chimera states for coupled pendula,” Scienti…c Reports 4, 6379 (2014)
[10] Keren Li, Shen Ma, Haihong Li, and Junzhong Yang, “Transition to synchronization in a kuramoto
model with the …rst- and second-order interaction terms,” Phys. Rev. E 89, 032917 (2014)

[11] Erik Andreas Martens, Shashi Thutupalli, Antoine Fourrière, and Oskar Hallatscheka, “Chimera
states in mechanical oscillator networks,” PNAS 110, 10563 (2013)
[12] Gautam C. Sethia, Abhijit Sen, and George L. Johnston, “Amplitude-mediated chimera states,”
Phys. Rev. E 88, 042917 (2013)

[13] Gautam C. Sethia and Abhijit Sen, “Chimera states: The existence criteria revisited,” Phys. Rev.
Lett. 112, 144101 (2014)
[14] Arkady Pikovsky, Michael Rosenblum, and Jürgen Kurths, Synchronization: A Universal Concept
in Nonlinear Sciences, 1st ed. (Cambridge University Press, Cambridge, 2003)
[15] Nan Yao, Zi-Gang Huang, Ying-Cheng Lai, and Zhi-Gang Zheng, “Robustness of chimera states in
complex dynamical systems,” Scienti…c Reports 3, 3522 (2013)
[16] Y. Zhu, Y. Li, M. Zhang, and J. Yang, “The oscillating two-cluster chimera state in non-locally
coupled phase oscillators,” Europhys. Lett. 97, 10009 (2012)
[17] A. Zakharova, M. Kapeller, and E. Scholl, “Chimera death: Symmetry breaking in dynamical net-
works,” Phys. Rev. Lett. 112, 154101 (2014)