FINITE-TIME ESTIMATOR-BASED ACTIVE CONTROL FOR TIME-DELAYED FRACTIONAL-ORDER NEURAL NETWORKS

Abstract

Time-delayed fractional-order systems are crucial in modeling and analyzing various physical systems, ranging from mechanical and electrical systems to biological and environmental ones. While estimators play an inevitable role in achieving high accuracy in controlling nonlinear systems, control techniques intended for time-delayed fractional-order systems struggle to estimate uncertainties within finite time. To address this issue, this study proposes a novel control technique that utilizes a finite-time disturbance observer and an active controller for time-delayed fractional-order systems. The stability of this method and the finite-time convergence of the estimator are guaranteed using the Lyapunov stability theorem and active control concepts. Then, the study investigates a fractional-order neural network and exhibits its chaotic behavior. Finally, the synchronization results of the fractional-order time-delayed neural network using the proposed control scheme in the presence of external disturbances are presented, verifying the effectiveness and robustness of the proposed control technique.