Exponential dissipativity of Cohen-Grossberg neural networks with mixed delays and reaction-diffusion terms
Baotong Cui · Control theory & applications · 2008
Without assuming the boundedness,monotonicity and differentiability of the activation functions and the boundedness of average time delays,the invariant set,global exponential stability and exponential dissipativity of a class of Cohen-Grossberg neural networks with mixed delays and reaction-diffusion terms are studied by employing the properties of diffusion operator,M-matrix and inequality technique.Some correlative sufficient conditions are then derived.The method used in this paper ignores the common requirement of constructing proper Lyapunov functionals,thus eliminat- ing the difficulty in building proper Lyapunov functionals.Moreover,the results extend and improve the prior results in publications.Finally,a numerical example is given to illustrate the effectiveness of the results.