Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
4947814 | Neurocomputing | 2017 | 8 Pages |
A theoretical investigation of neutral-type of delayed inertial neural networks using the Lyapunov stability theory and Linear Matrix Inequality (LMI) approach is presented. Based on a suitable variable transformation, an inertial neural network consisting of second-order differential equations can be converted into a first-order differential model. The sufficient conditions of the delayed inertial neural network are derived by constructing suitable Lyapunov functional candidates, introducing new free weighting matrices, and utilizing the Writinger integral inequality. Through the LMI solution, we analyse the global asymptotic stability condition of the resulting delayed inertial neural network. Simulation examples are presented to demonstrate the effectiveness of the derived analytical results.