Interval time-varying delay stability for neural networks

This paper presents a new linear matrix inequality (LMI) stability criterion for continuous-time artificial neural networks (ANN) with interval time-varying delay. The varying-time delay is taken as composition of a nominal positive value subject to a time-varying perturbation. The methodology is based on Gu's discretization technique and a strategy that decouples the system matrices from the Lyapunov functional matrices. Two numerical examples are performed to support the theoretical predictions.