Dual-mechanism adaptive control for finite/fixed-time synchronization of fuzzy inertial neural networks under parameter uncertainty.

Abstract

This paper investigates finite/fixed-time synchronization of fuzzy inertial neural networks (FINNs) under parameter uncertainty via two complementary adaptive control mechanisms. Rather than constructing a single controller, the study work separately designs and analyzes adaptive synchronization controllers based on both reduced and non-reduced methods, taking into account the inertial dynamics, fuzzy nonlinearities, and time-varying delays in FINNs. A non-reduced approach is developed by retaining the full inertial dynamics to ensure high synchronization accuracy, while a reduced approach transforms the system into simplified dynamics for efficient implementation. Leveraging Lyapunov stability theory and finite/fixed-time stability lemmas, we derive synchronization criteria that ensure faster convergence and enhanced real-time performance in finite-time synchronization, while fixed-time synchronization further guarantees uniform convergence within a predictable time, independent of initial conditions. Comparative simulations demonstrate that the non-reduced method achieves higher precision by retaining full inertial dynamics, whereas the reduced method leverages structural simplification and order reduction to facilitate faster convergence and improved computational efficiency. These results establish a dual-mechanism framework for robust time-constrained synchronization of uncertain FINNs.