一类递归神经网络系统的有限时间同步

Finite-time Synchronization for a Class of Recurrent Neural Network System

针对非线性电路网络系统中的特殊系统—递归神经网络系统,本文拟采用事件触发间歇控制策略研究递归神经网络的有限时间同步问题.首先,根据有限时间间歇控制的理论推导过程,提出事件触发间歇控制规则;其次,提出间歇控制器,利用Lyapunov稳定性方法以及一些不等式技巧,得到了实现驱动-响应神经网络系统的有限时间同步的充分条件,并排除了Zeno行为;最后,通过数值实例验证了所提控制策略的有效性.

Aiming at the special system of nonlinear circuit network system,that is recurrent neural network system,the finite time synchronization of recurrent neural network by using event triggered intermittent control strategy is study in this paper.Firstly,the event triggered intermittent control rules are proposed according to the theoretical derivation process of finite time intermittent control;Secondly,the intermittent controller is presented.The sufficient conditions for realizing the finite time synchronization of drive-response neural network system are obtained,and the Zeno behavior is excluded using the method of Lyapunov stability and some inequality techniques;Finally,the numerical example is given to verify the effectiveness of the proposed control strategy.