Non-fragile sampled-data control for synchronization of chaotic fractional-order delayed neural networks via LMI approach

This study addresses the master and slave synchronization problem of chaotic fractional-order delayed neural networks using a non-fragile sampled-data control (NFSDC) scheme which includes uncertainty information in the control gain matrix. An appropriate Lyapunov functional is constructed with information about sampling instants. A new, improved fractional-order inequality is developed to estimate the integral term. Then, based on the new integral inequality, the delay-dependent stability criteria are derived in the form of linear matrix inequality, which guarantees the asymptotic stability of the fractional-order error systems. This implies that the proposed NFSDC can synchronize the fractional-order master and slave systems. Moreover, numerical simulation results illustrate the synchronization nature and the influence of fractional derivatives in the system dynamics. Finally, it can be concluded that the proposed method and control scheme are effective.

Automated summary

This study proposes a non-fragile sampled-data control scheme to solve the master and slave synchronization problem of chaotic fractional-order delayed neural networks. By incorporating uncertainty information in the control gain matrix and using an appropriate Lyapunov functional, the delay-dependent stability criteria are derived in the form of linear matrix inequality. The proposed scheme successfully synchronizes the fractional-order master and slave systems, as demonstrated by numerical simulations. Overall, the method and control scheme are effective.