Key Words:FINITE-TIME; SYNCHRONIZATION; BIFURCATIONS; STABILITY
Abstract:Most of the complex network in the real world are not single-layer networks, and networks will be connected with each other. Networks with multi-layer is important because it means cognitive and artificial intelligence. Most current studies of networks consider the case that with n-nodes including ring network, small-word network, scale-free network, etc. This type of network is not enough to describe the complex structure of actual neural networks. However, it is more actual to study the dynamic behavior of multi-layer networks than single-layer networks. In this paper, the stability and bifurcation of a class of three-layer fractional-order neural networks with multiple delays was studied for the first time. By selecting the appropriate bifurcation parameter, the internal dynamic behavior of the given model was discussed by using the theory of Hopf bifurcation, and the critical value and criterion for Hopf bifurcation are derived. The influence of delay, fractional order, and the number of hidden neurons on the bifurcation point were discussed in detail. And the critical value of Hopf bifurcation is accurately calculated. The results show that the stability of the system can be destroyed by increasing the fractional order and the number of hidden neurons. The correctness of the theoretical results is verified by numerical simulation.
Volume:17
Issue:1
Translation or Not:no