具反应扩散项变时滞细胞神经网络模型指数稳定性的新结果  

A New Criterion on the Global Exponential Stability for Reaction-Diffusion Cellular Neural Networks with Time-Varying Delays

作　　者：罗毅平[1] 夏文华[1] 刘国荣[1] 邓飞其[2] 

机构地区：[1]湖南工程学院,湖南湘潭411101 [2]华南理工大学系统工程研究所,广东广州510640

出　　处：《电子学报》2008年第4期609-613,共5页Acta Electronica Sinica

基　　金：国家自然科学基金(No.60374023);湖南省自然科学基金(No.07JJ6112);湖南省教育厅重点项目(No.04A012,No.07A015)

摘　　要：针对一类具反应扩散项的变时滞细胞神经网络模型,利用Poincare不等式与Hanalay不等式等知识,获得了该系统的指数稳定性条件,该稳定性条件包含了扩散算子项,与以往结果比较,获得的指数稳定性条件更强,且降低了已有结论的保守性.最后,通过实例说明该方法的有效性.Based on extended Hanalay inequality, the exponential asymptotic stability for a class of cellular neural networks with delays and reaction-diffusion terms is investigated by using Poincare inequality. The obtained criteria have a distinguished feature from previous studies, and our results contain diffusion operator terms. The new results are less conservative than the existing ones. Finally, an illustrative example is give to verify the effectiveness of the method.

关 键 词：细胞神经网络 反应扩散 指数稳定性 变时滞 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]