混合时变时滞随机神经网络系统的耗散控制

发布时间：2018-05-29 02:50

本文选题：遗忘混和时变时滞随机神经网络系统 + 全局均方稳定　；参考：《辽宁大学》2017年硕士论文

【摘要】：人工神经网络具有自组织、自学习、快速找寻最优化解的能力,并在模式识别、人工智能、自动控制、通信和物流等领域有广泛应用.本文主要研究混和时变时滞随机神经网络系统的耗散控制问题.主要内容如下:首先,介绍神经网络的历史发展及系统概况,给出本文的结构与安排;其次,研究遗忘随机神经网络的全局均方稳定的问题,运用随机稳定性理论和线性矩阵不等式等方法,得到系统全局均方稳定的充分条件,并设计状态估计器使误差系统渐近稳定,通过两个数值算例验证结论的可行性;再次,分别对离散和连续混合时变时滞随机神经网络系统进行γ耗散、指数耗散和指数无源研究,运用LMI方法,得到系统γ耗散、指数耗散和指数无源的充分条件,通过数值算例验证结论的可行性;最后对本文进行总结.
[Abstract]:Artificial neural network has the ability of self-organization, self-learning and fast searching for optimal solution, and it has been widely used in the fields of pattern recognition, artificial intelligence, automatic control, communication and logistics. In this paper, the dissipative control problem of stochastic neural network systems with mixed time-varying delays is studied. The main contents are as follows: firstly, the history and system of neural networks are introduced, and the structure and arrangement of this paper are given. Secondly, the problem of global mean square stability of forgotten stochastic neural networks is studied. By using stochastic stability theory and linear matrix inequality (LMI), the sufficient conditions for global mean square stability of the system are obtained, and a state estimator is designed to make the error system asymptotically stable. The feasibility of the conclusion is verified by two numerical examples. The 纬 dissipation, exponential dissipation and exponential passivity of discrete and continuous hybrid time-varying delay stochastic neural networks are studied, respectively. By using LMI method, sufficient conditions of 纬 dissipation, exponential dissipation and exponential passivity are obtained. The feasibility of the conclusion is verified by numerical examples. Finally, this paper is summarized.
【学位授予单位】：辽宁大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：O175

【参考文献】