Network-based analysis of fluid flows: Progress and outlook

Abstract

The network of interactions among fluid elements and coherent structures gives rise to the incredibly rich dynamics of vortical flows. These interactions can be described with the use of mathematical tools from the emerging field of network science, which leverages graph theory, dynamical systems theory, data science, and control theory. The blending of network science and fluid mechanics facilitates the extraction of the key interactions and communities in terms of vortical elements, modal structures, and particle trajectories. Phase-space techniques and time-delay embedding enable a network-based analysis of time-series measurements in terms of visibility, recurrence, and cluster transitions. Equipped with the knowledge of interactions and communities, the network-theoretic approach enables the analysis, modeling, and control of fluid flows, with a particular emphasis on interactive dynamics. In this article, we provide a brief introduction to network science and an overview of the progress on network-based strategies to study the complex dynamics of fluid flows. Case studies are surveyed to highlight the utility of network-based techniques to tackle a range of problems from fluid mechanics. Towards the end of the paper, we offer an outlook on network-inspired approaches.