Synchronization in the Kuramoto model: a dynamical gradient network approach.

We propose a dynamical gradient network approach to consider the synchronization in the Kuramoto model. Our scheme to adaptively adjust couplings is based on the dynamical gradient networks, where the number of links in each time interval is the same as the number of oscillators, but the links in different time intervals are also different. The gradient network in the (n+1)th time interval is decided by the oscillator dynamics in the n th time interval. According to the gradient network in the (n+1)th time interval, only one inlink's coupling for each oscillator is adjusted by a small incremental coupling. During the transition to synchronization, the intensities for all oscillators are identical. Direct numerical simulations fully verify that the synchronization in the Kuramoto model is realized effectively, even if there exist delayed couplings and external noise.