Solitons are special analytic solutions to the nonlinear wave equations that turn up everywhere: from fluid dynamics to quantum mechanics to molecular biology. One such equation is the Korteweg–de Vries (KdV) equation for a 1D wave $u(x,t)$ evolving in time $t$: \[ \frac{\partial u}{\partial t}+6u\frac{\partial u}{\partial x}+\frac{\partial^3 u}{\partial x^3}=0. \] What makes solitons special is that they behave in many ways like solutions to the linear wave equation: propagating without losing their shape, and even interacting nicely:
In 1990, Daisuke Takahashi and Junkichi Satsuma proposed a discrete model for solitons called the ball and box model. It consists of an infinite row of boxes, some of which contain a ball. A sequence of consecutive balls represents a wave. The pattern of waves after each time step is found by starting from the left and moving each ball to the next available empty box to its right.
A row of $n$ consecutive balls behaves like a single wave, and moves to the right at a constant speed $n$. Several such waves, so long as they start sufficiently separated, will interact (in a possibly messy way), and then disperse, maintaining their original shapes overall.
This model is fun to play with, but—surprisingly given its simplicity—it manages to capture a large amount of the interesting and unusual behaviour of the fluid waves modelled by the KdV equation.
More from Chalkdust
Ode to Newtonian mechanics
(after Sonnet 130)
My favourite LaTeX package
The Chalkdust editors share some of their favourites
In conversation with Ulrike Tillmann
Sophie Maclean and David Sheard speak to a very top(olog)ical mathematician!
Dear Dirichlet, Issue 12
Books, business and barns are the topics readers have sent in to the professor's postbox this issue
Prize crossnumber, Issue 12
£100 of MathsGear goodies to be won if you can solve it
#MarkovMadeMeTweet
A handy tool that can make writing great tweets much easier