论文标题

库拉莫托网络中同步的子模块输入选择

Submodular Input Selection for Synchronization in Kuramoto Networks

论文作者

Sahabandu, Dinuka, Clark, Andrew, Bushnell, Linda, Poovendran, Radha

论文摘要

同步是工程和自然网络动力学系统的重要属性。非线性同步的Kuramoto模型已在应用中广泛研究,包括夹带脑网络中的时钟细胞和电力系统稳定性。在振荡器之间存在签名耦合的情况下,当网络包括具有异质固有频率的振荡器时,库拉莫托网络的同步在存在签名耦合的情况下具有挑战性。在本文中,我们研究了最小设定控制输入选择的问题,用于同步签名的库拉莫托网络。我们首先使用基于被动率的框架来得出足够的条件,以在均质和异质库拉莫托网络中同步。然后,我们开发了一种用于选择给定库拉莫托网络的最小控制输入集的算法。我们通过一项数值研究对多种图形(包括无向,定向和周期图)进行数字研究来评估我们的方法。

Synchronization is an essential property of engineered and natural networked dynamical systems. The Kuramoto model of nonlinear synchronization has been widely studied in applications including entrainment of clock cells in brain networks and power system stability. Synchronization of Kuramoto networks has been found to be challenging in the presence of signed couplings between oscillators and when the network includes oscillators with heterogeneous natural frequencies. In this paper, we study the problem of minimum-set control input selection for synchronizing signed Kuramoto networks. We first derive sufficient conditions for synchronization in homogeneous as well as heterogeneous Kuramoto networks using a passivity-based framework. We then develop a submodular algorithm for selecting a minimum set of control inputs for a given Kuramoto network. We evaluate our approach through a numerical study on multiple classes of graphs, including undirected, directed, and cycle graphs.

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