학술논문
Analysis on Robust Passivity of Uncertain Neural Networks with Time-Varying Delays via Free-Matrix-Based Integral Inequality
Document Type
Article
Source
(2022): 2385-2394.
Subject
Language
Korean
ISSN
15986446
Abstract
This paper investigates the robust delay-dependent passivity problem of neural networks (NNs) with time-varying delays and parameter uncertainties. A suitable augmented Lyapunov-Krasovskii functional (LKF) with triple integral term, which takes full use of the neuron activation function conditions and the information of time-delay in integral term, is constructed. Furthermore, by utilizing integral inequality proposed recently and the combining reciprocally convex method to estimate the derivative of the LKF, some less conservative robust passivity conditions are derived in terms of LMI. The superiority of presented approaches are demonstrated via two classic numerical examples.
This paper investigates the robust delay-dependent passivity problem of neural networks (NNs) with time-varying delays and parameter uncertainties. A suitable augmented Lyapunov-Krasovskii functional (LKF) with triple integral term, which takes full use of the neuron activation function conditions and the information of time-delay in integral term, is constructed. Furthermore, by utilizing integral inequality proposed recently and the combining reciprocally convex method to estimate the derivative of the LKF, some less conservative robust passivity conditions are derived in terms of LMI. The superiority of presented approaches are demonstrated via two classic numerical examples.
This paper investigates the robust delay-dependent passivity problem of neural networks (NNs) with time-varying delays and parameter uncertainties. A suitable augmented Lyapunov-Krasovskii functional (LKF) with triple integral term, which takes full use of the neuron activation function conditions and the information of time-delay in integral term, is constructed. Furthermore, by utilizing integral inequality proposed recently and the combining reciprocally convex method to estimate the derivative of the LKF, some less conservative robust passivity conditions are derived in terms of LMI. The superiority of presented approaches are demonstrated via two classic numerical examples.