时滞细胞神经网络全局同步的滑模控制方法  

Global synchronization of cellular neural network with time-delays by modified sliding mode control method

作　　者：蔡国梁[1] 姚琴[1,2] 姜胜芹 

机构地区：[1]江苏大学非线性科学研究中心,江苏镇江212013 [2]江苏物联网研究发展中心,江苏无锡224000

出　　处：《江苏大学学报（自然科学版）》2014年第3期366-372,共7页Journal of Jiangsu University：Natural Science Edition

基　　金：国家自然科学基金资助项目(51276081;91010011);教育部社科基金资助项目(12YJAZH002);江苏高校优势学科建设工程项目(1033000001);江苏大学高级人才基金资助项目(07JD054);江苏大学学生科研项目(10A147;12A415)

摘　　要：利用改进的滑模控制方法,研究了具有多时滞和分布时滞细胞神经网络的全局同步问题.基于Lyapunov稳定性原理,运用线性矩阵不等式方法,设计滑模面,考虑了时滞独立和时滞相关两种情况.设计了合适的滑模控制器,以使得误差系统的轨迹能全局达到滑模面.得到了相对充分的条件,来确保时滞独立与时滞相关两种情况下误差系统的全局稳定性,从而实现时滞细胞神经网络的全局同步.结果表明所设计的滑模控制器由误差反馈控制输出来获得高增益补偿的准则,可用以处理神经网络的全局同步问题,并且此方法可应用于其他复杂神经网络.Using the modified sliding mode control method, the global synchronization of cellular neural network was investigated with multi-delays and distributed delays. Based on the Lyapunov stability theorem, by the linear matrix inequality technique and designing a suitable sliding surface, a sufficient condition was derived to guarantee the global stability of the error system under delay-independent and delaydependent situations. The feasible sliding mode control law was designed so that the trajectory of the error system was globally driven onto the specified sliding surface. The global synchronization of cellular neural network with both multi-delays and distributed delays was realized. The results show that the proposed sliding mode controller can be designed to achieve high gain compensation by the error output feedback control, and can be used to deal with the global synchronization of neural networks. The proposed method can also be applied for other complex neural networks.

关 键 词：时滞细胞神经网络 滑模控制 全局稳定 全局同步 延迟 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置] N93[自动化与计算机技术—控制科学与工程]