Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
4974581 | Journal of the Franklin Institute | 2015 | 19 Pages |
Abstract
This paper focuses on the problems of stability and dissipativity analysis for static neural networks (NNs) with interval time-varying delay. A new augmented Lyapunov-Krasovskii functional is firstly constructed, in which the information on the activation function is taken fully into account. Then, by employing a Wirtinger-based inequality to estimate the derivative of Lyapunov-Krasovskii functional, an improved stability criterion is derived for the considered neural networks. The result is extended to dissipativity analysis and a sufficient condition is established to assure the neural networks strictly dissipative. Two numerical examples are provided to demonstrate the effectiveness and the advantages of the proposed method.
Related Topics
Physical Sciences and Engineering
Computer Science
Signal Processing
Authors
Hong-Bing Zeng, Ju H. Park, Chang-Fan Zhang, Wei Wang,