Czajkowski Bruno M, Viana Ricardo L
Departamento de Física, Universidade Federal do Paraná, 81531-990 Curitiba, Paraná, Brazil.
Universidade Federal do Paraná, Centro Interdisciplinar de Ciência, Tecnologia e Inovação, Núcleo de Modelagem e Computação Científica, Curitiba-PR, Brazil.
Chaos. 2024 Sep 1;34(9). doi: 10.1063/5.0219961.
Unstable dimension variability is an extreme form of non-hyperbolic behavior that causes a severe shadowing breakdown of chaotic trajectories. This phenomenon can occur in coupled chaotic systems possessing symmetries, leading to an invariant attractor with riddled basins of attraction. We consider the coupling of two Lorenz-like systems, which exhibits chaotic synchronized and anti-synchronized states, with their respective basins of attraction. We demonstrate that these basins are riddled, in the sense that they verify both the mathematical conditions for their existence, as well as the characteristic scaling laws indicating power-law dependence of parameters. Our simulations have shown that a biased random-walk model for the log-distances to the synchronized manifold can accurately predict the scaling exponents near blowout bifurcations in this high-dimensional coupled system. The behavior of the finite-time Lyapunov exponents in directions transversal to the invariant subspace has been used as numerical evidence of unstable dimension variability.
不稳定维度变异性是一种非双曲行为的极端形式,它会导致混沌轨迹的严重跟踪失效。这种现象可能发生在具有对称性的耦合混沌系统中,从而导致具有满是孔洞的吸引盆的不变吸引子。我们考虑两个类洛伦兹系统的耦合,它们表现出混沌同步和反同步状态以及各自的吸引盆。我们证明这些吸引盆是满是孔洞的,即它们既满足其存在的数学条件,也满足表明参数幂律依赖性的特征标度律。我们的模拟表明,对于到同步流形的对数距离的有偏随机游走模型,可以准确预测这个高维耦合系统中接近爆裂分岔处的标度指数。与不变子空间横向方向上的有限时间李雅普诺夫指数的行为已被用作不稳定维度变异性的数值证据。