Finite-Time Bipartite Synchronization of Delayed Antagonistic Memristive Neural Networks: Optimizing Event-Triggered Controller Parameters

This article is concerned with the finite-time bipartite synchronization problem of delayed antagonistic memristive neural networks through controller design. A tractable error system is established, where the coordinate transformation technique and interval matrix method are employed to handle cooperation-competition interactions and the memristive connection weights, respectively. A coopetition-dependent event-triggered controller is constructed, which can reduce network bandwidth utilization and eliminate Zeno behavior. Taking into account the residual terms, the system delay, the finite-time constraint, and the influence of cooperation-competition interactions, some new control parameters are introduced into the designed controller. By resorting to the inequality techniques and the Lyapunov theory, a finite-time bipartite synchronization criterion is established. Then, within the established theoretical framework, an optimization model is designed to achieve bipartite synchronization in minimal time. Accordingly, a sparrow search-based optimization algorithm is employed to derive the optimal controller parameters. Finally, effectiveness and advantages of this research are validated through both a numerical example and a practical application.