An independently attributed heterogeneous-model chaos report

Popovych, Maistrenko and Tass report chaotic behavior in a mathematical model of four globally coupled phase oscillators with nonidentical natural frequencies. Their equations are ${\dot{\theta }}_i={\omega }_i+\frac{K}{4}{\sum }_{j=1}^4\sin ({\theta }_j-{\theta }_i)$, where ${\omega }_i=-1+2(i-1)\mathrm{/}3$ for $i=1,2,3,4$. A phase is an angle modulo $2\pi$; ${\omega }_i$ is the constant rate of the $i$th phase when $K=0$, and $K$ is the common coupling parameter. Thus the natural frequencies are $-1,-1\mathrm{/}3,1\mathrm{/}3,1$.

The authors reduce the dynamics to the three phase differences ${\varphi }_i={\theta }_{i+1}-{\theta }_i$ and report a chaotic attractor: an attracting long-term behavior exhibiting chaos in their numerical study. They report a positive maximal Lyapunov exponent, a diagnostic of exponential separation of nearby trajectories, and display a section of the attractor. They associate the transition with torus destruction. This is a report about a mathematical model, not a measurement of physical oscillators or a locally verified rigorous existence theorem. Separately, the abstract of Maistrenko, Popovych and Tass's paper Chaotic Attractor in the Kuramoto Model reports a chaotic attractor in a four-dimensional Kuramoto model and studies its origin.

Finite phase tori, local lifts and the flat metric

Phasors, trigonometric identities and periodic derivatives

The reported model, frequency vector, reduction and diagnostics are imported from Popovych, Maistrenko and Tass, Phase chaos in coupled oscillators, equation (1), the four-oscillator discussion and Figure 1. The authors report the maximal exponent as positive approximately for $0.94<K<1.22$; these endpoints and the chaos identification are their numerical findings, not a locally reproduced computation. The Poincaré section shown is the set ${\varphi }_1=0$ in the reduced phase space, used to display successive intersections of trajectories with that set. A Lyapunov exponent describes the long-time exponential growth rate of an infinitesimal perturbation along a trajectory; the maximal exponent is the largest such rate. A torus-destruction scenario refers here to the authors' reported transition from motion associated with a torus to the chaotic behavior they display. The separate report attributed to Chaotic Attractor in the Kuramoto Model is limited to its abstract.

The numerical observations and their interpretation are attributed to the cited authors. This item does not import a proof that a positive numerical exponent alone establishes a mathematically defined chaotic attractor, nor does it assert that the two cited papers use the same parameter vector.