Synchronization of coupled memristive inertial delayed neural networks with impulse and intermittent control
- PDF / 619,333 Bytes
- 12 Pages / 595.276 x 790.866 pts Page_size
- 13 Downloads / 211 Views
(0123456789().,-volV)(0123456789().,-volV)
ORIGINAL ARTICLE
Synchronization of coupled memristive inertial delayed neural networks with impulse and intermittent control Wei Zhang1
•
Jiangtao Qi2
Received: 13 March 2020 / Accepted: 13 November 2020 Ó Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract This paper contributes to the global exponential synchronization under impulses effective and periodically intermittent control for linearly coupled memristive inertial delayed neural networks (MIDNNs). First of all, we built an array of linearly coupled MIDNNs. Second, by using linear matrix inequality, Lyapunov function, comparison principle, and an extended Halanay differential inequality, we derived some conclusions which rely on impulses effects and periodically intermittent control insures the global exponential synchronization of the coupled MIDNNs. In the end, two instances put forward illustrate the feasibility of the theoretical results. Keywords Time-varying delays Coupled MIDNNs Periodically intermittent control Impulse control
1 Introduction Memristor has been widely investigated in the last few decades, from Chua [1] to Hewlett-Packard [2] to make physical objects. Memristors have many of the characteristics of resistance, such as the same measurement unit, and provide nonvolatile storage in a simple device structure. In addition, it has the characteristics of nanometer, low power consumption and strong storage capacity and has broad application prospects in circuit design and optimization and so on. Its main feature is the unique ability to remember the amount of electric charge in the past. On this basis, by using a memristor substitute a resistor, build a new neural networks (NNs) model to simulate the human brain, and hope to apply the memristor to the next-generation computer. Artificial memristive neural network (MNN) is one
& Wei Zhang [email protected] Jiangtao Qi [email protected] 1
Chongqing Key Laboratory of Nonlinear Circuits and Intelligent Information Processing, Department of Electronics and Information Engineering, Southwest University, Chongqing 400715, China
2
School of Information Science and Electrical Engineering, Shan Dong Jiaotong University, Jinan 250357, China
of the important applications, and it shows abundant of dynamical phenomenon, for synchronization, instance periodic solutions, consensus, stability, etc. [3–8]. In [3], based on theories of differential inclusion and Filippov solution, exponential adaptive lag synchronization has been investigated for MNNs. In [7], the authors have been derived globally exponential stability of MDNNs. In [8], by applying new Lyapunov function, the authors not only have been researched the existence and uniqueness, but also have been studied exponential stability of the almost periodic MDNNs with leakage, which the delay is distributed delays and time varying. As is known to all, many NNs studied the first-order states, such as cell NNs, Lotka–Volterra NNs, Hopfield NNs, recurrent NNs and so on. How
Data Loading...
