Synchronization of Markov jump neural networks with two delay components via Affine transformed sampled-data control with actuator saturation

• Published in: European physical journal plus Vol. 139; no. 10; p. 913
• Main Authors: Subhashri, A. R., Radhika, T.
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 21.10.2024; Springer Nature B.V
• Subjects: Actuators; Affine transformations; Applied and Technical Physics; Atomic; Communication; Complex Systems; Condensed Matter Physics; Control algorithms; Controllers; Dynamical systems; Functionals; Linear matrix inequalities; Markov analysis; Mathematical and Computational Physics; Molecular; Neural networks; Optical and Plasma Physics; Parameter modification; Physics; Physics and Astronomy; Regular Article; Signal processing; State vectors; Stochastic models; Theoretical; Time lag; Time synchronization; Time varying control; Upper bounds; Weighting functions
• ISSN: 2190-5444; 2190-5444
• DOI: 10.1140/epjp/s13360-024-05695-x

In this paper, in contrast to the existing findings the synchronization criteria for chaotic neural networks (CNNs) with sampled data control is analysed by a new integral inequality. The proposed method incorporates a parameterized controller that depends on the activation function and includes actuator saturation in addition to Markovian jump CNNs (MJCNNs) and additive time varying delay. The reformulated approach considers the constraints on the activation function parameters using weighting functions. Additionally, the controller gain matrices are combined using weighted functions that undergo affine transformations. The construction of Lyapunov–Krasovskii functionals (LKFs) lead to the establishment of two augmented terms, which facilitate the interaction among the state vectors with upper bounds of additive time delay. Benefitting from the modified free matrix-based integral inequalities addressed in Lemmas  1 and 2 provide the sufficient conditions for the affine transformed controller of the MJCNNs error system in the form of linear matrix inequalities (LMIs). Further, simulation examples are provided to demonstrate the effectiveness and superiority of the affine transformed control approach.