Finite-Time Stabilization of Semi-Markov Reaction-Diffusion Memristive NNs with Unbounded Time-Varying Delays

Jun Zhang,Song Zhu,Kai-Ning Wu,Mouquan Shen,Shiping Wen

DOI: https://doi.org/10.1109/tcsi.2024.3459913

2024-01-01

Abstract:This paper mainly analyzes the finite-time stabilization of semi-Markov reaction-diffusion memristive neural networks (R-DMNNs) with unbounded time-varying delays. Firstly, the reaction-diffusion term and semi-Markov jumping are introduced into memristive neural networks, which relaxes the limitation of Markov switching on sojourn time and makes the model more applicable. Secondly, by constructing a suitable comparison function, the states of R-DMNNs converges to $0$ directly, which can clearly estimate the upper limit of the settling time and simplify the complexity of the theoretical derivation. Furthermore, this paper removes the requirement of bounded and differentiable time delay, which provides a new perspective for understanding the finite-time stabilization of the neural networks with reaction-diffusion terms. Finally, one example illustrates the usefulness of the analysis results in this research.