论文标题

部分可观测时空混沌系统的无模型预测

Emergence of Self-dual Patterns in Active Colloids with Periodical Feedback to Local Density

论文作者

Yang, Yang, Zhang, Zhi Chao, Qi, Fei, Zhang, Tian Hui

论文摘要

储层计算是预测湍流的有力工具,其简单的架构具有处理大型系统的计算效率。然而,其实现通常需要完整的状态向量测量和系统非线性知识。我们使用非线性投影函数将系统测量扩展到高维空间,然后将其输入到储层中以获得预测。我们展示了这种储层计算网络在时空混沌系统上的应用,该系统模拟了湍流的若干特征。我们表明,使用径向基函数作为非线性投影器,即使只有部分观测并且不知道控制方程,也能稳健地捕捉复杂的系统非线性。最后,我们表明,当测量稀疏、不完整且带有噪声,甚至控制方程变得不准确时,我们的网络仍然可以产生相当准确的预测,从而为实际湍流系统的无模型预测铺平了道路。

The central task in the study of self-organization is to explore the general mechanism of emergences. However, this is inhibited by the missing of a full knowledge of the microscopic dynamics of emergence. Here, in this study, the microscopic dynamics of self-organization for patterns is investigated and quantified in a periodically propelled Quincke system. The periodical coupling between propulsion and repulsion at the particle level leads to local directed oscillating particle flows and promises a loop of positive feedback to density fluctuations. Nevertheless, the global evolution of the resulting cluster phase is dominated by a global dual transformation. As stable attractors of the dual transformation, self-dual patterns including stripe patterns and square lattices can be achieved by tuning the strength and the frequency of propelling. However, stripes are possible only at strong propelling where boundary particle flows can form. The findings in this study show that the dynamics of emergence on different length scales are controlled by different mechanisms. The competition and the interplay between different microscopic dynamic processes play the central role in determining the product of emergence. Moreover, the periodically oscillating self-dual patterns demonstrate a classical approach to time crystals.

扫码加入交流群

加入微信交流群

微信交流群二维码

扫码加入学术交流群,获取更多资源