Open Access

ARTICLE

Delay-dependent Stability of Recurrent Neural Networks with Time-varying Delay

Guobao Zhang^a,b, Jing-Jing Xiong^a,b, Yongming Huang^a,b, Yong Lu^a,b,c, Ling Wang^a,b

^a School of Automation, Southeast University, Nanjing 210096, China
^b Key Laboratory of Measurement and Control of CSE (School of Automation, Southeast University), Ministry of Education, Nanjing 210096, China
^c Earthquake Administration of Jiangsu Province, Nanjing 210014, China

* Corresponding Author: Yongming Huang,

Intelligent Automation & Soft Computing 2018, 24(3), 541-551. https://doi.org/10.31209/2018.100000021

Abstract

This paper investigates the delay-dependent stability problem of recurrent neural networks with time-varying delay. A new and less conservative stability criterion is derived through constructing a new augmented Lyapunov-Krasovskii functional (LKF) and employing the linear matrix inequality method. A new augmented LKF that considers more information of the slope of neuron activation functions is developed for further reducing the conservatism of stability results. To deal with the derivative of the LKF, several commonly used techniques, including the integral inequality, reciprocally convex combination, and free-weighting matrix method, are applied. Moreover, it is found that the obtained stability criterion has a lower computational burden than some recent existing ones. Finally, two numerical examples are considered to demonstrate the effectiveness of the presented stability results.

---

Keywords

---