摘要
对一类具变时滞脉冲Hopfield神经网络的周期解进行研究,输出函数满足Lipschitz条件,但不必是连续可微的。不建立合适的Lyapunov泛函,而是利用迭代和不等式技巧,得到该神经网络周期解的存在性与一致稳定性的一个充分条件,该条件简单且容易验证。数值算例及仿真结果验证了所得结论的正确性与有效性。
The periodic solution of impulsive Hopfield neural networks with time-varying delay is studied. The output function satisfies Lipschitz condition, but is not necessarily continuous differentiable. Iterative analysis and inequality techniques instead of Lyapunov functional are used. A sufficient condition for the existence and uniform stability of periodic solution of the system is obtained. This condition is simple and easy to verify. An numerical example and its simulation are given to illustrate the correctness and effec- tiveness of the obtained results.
出处
《黑龙江大学自然科学学报》
CAS
北大核心
2016年第2期156-161,共6页
Journal of Natural Science of Heilongjiang University
基金
国家自然科学基金资助项目(61374009)
关键词
HOPFIELD神经网络
变时滞
周期解
一致稳定性
脉冲干扰
Hopfield neural networks
time-varying delays
periodic solution
uniform stability
impulsive perturbation
