一类非线性输入时滞系统自适应控制——无源化方法  

Adaptive control for a class of input-delay nonlinear systems by a passive approach

作　　者：于占东[1] 王显峰[1] 

机构地区：[1]哈尔滨工程大学自动化学院,黑龙江哈尔滨150001

出　　处：《哈尔滨工程大学学报》2011年第2期188-193,共6页Journal of Harbin Engineering University

基　　金：国家自然科学基金资助项目(60804009;60704004);教育部高等学校博士点专项基金资助项目(20070217034);哈尔滨工程大学基础研究基金资助项目(HEUFT05062)

摘　　要：针对一类含输入时滞的不确定严反馈非线性系统,设计了有记忆时滞依赖型γ-无源自适应控制器.首先,在局部线性化的基础上设计了有记忆时滞依赖型γ-无源控制器.针对严反馈非线性系统各个子系统中的非线性,利用Back-stepping方法将局部γ-无源控制律进行分解,得到线性的中间虚拟控制律,在此基础上,利用神经网络补偿各子系统的非线性部分,将上述过程整合即得到系统的最终控制律.控制器的特点是针对系统中的输入时滞,反馈控制律采用有记忆的时滞依赖型控制策略,即系统的反馈控制律不仅与当前的系统状态有关,还与系统的时滞过程中的控制作用有关.这种控制方法比无记忆时滞独立型控制保守性更小.稳定性分析中,证明了闭环系统是一致终态有界稳定的.The γ-passivity delay-dependent control approach was presented for uncertain strict-feedback nonlinear systems with input-delay.First,the delay-dependent γ-passivity controller for the local linearization section of the nonlinear model was designed.Then,a γ-passivity control law of a local linear model was decomposed as the virtual control of sub-systems by the backstepping approach.In order to compensate for the nonlinear dynamics,an adaptive neural model was proposed.The design procedure of whole systems was regarded as a combination of a local γ-passivity control and adaptive neural network compensation techniques.One important feature of the presented scheme is that the controller with input history feedback had a lower conservation level than the ones without memory.Stability analysis implies that the closed-loop system is ultimately uniformly bounded.

关 键 词：严反馈非线性系统 输入时滞 γ-无源 BACKSTEPPING方法 神经网络 

分 类 号：TP270[自动化与计算机技术—检测技术与自动化装置]