A data-driven model-free adaptive pinning synchronization control study for complex networks

This paper explores the problem of synchronous control of discrete complex network dynamics models. In view of the challenges such as the difficulty of modeling complex networks, the complexity of network structure and the difficulty of controller design, this paper proposes an improved model-free adaptive pinning control method. First, a method of entropy of the betweenness centrality and node strength is proposed to select the key nodes, construct the augmentation and generalization error system, and design the control strategy based on the node input and output data. Second, the synchronous stability is analyzed theoretically and the controller parameters are optimized by firefly optimization algorithm in order to overcome the parameter tuning difficulties. Finally, the effectiveness of the proposed pinning node selection strategy in this paper is verified by simulation, and it is verified by simulation experiments of BA scale-free network and ER stochastic network that the pinning control method in this paper only needs to control a few key nodes in the network to realize the synchronous state of the whole network. The method of this paper provides a new idea for synchronous control of complex networks.