基于神经网络的一类非线性时滞系统自适应控制器设计

针对一类带有扰动和未知时滞的非线性系统,通过反步方法设计一种鲁棒自适应控制器.提出了一种新的Lyapunov-Krasovskii泛函,补偿了未知时滞项的不确定性.引入一种合适的偶函数,避免了控制器的奇异性问题.通过Lyapunov直接方法,证明了所设计的控制器能保证闭环系统所有信号全局一致最终有界.     

一类非线性时滞混沌系统的自适应脉冲同步

针对一类非线性时滞混沌系统,提出了一种新的自适应脉冲同步方案.首先基于Lyapunov稳定性理论、自适应控制理论及脉冲控制理论设计了自适应控制器、脉冲控制器及参数自适应律,然后利用推广的Barbalat引理,理论证明响应系统与驱动系统全局渐近同步,并给出了相应的充分条件.方案利用参数逼近Lipschitz常数,从而取消了Lipschitz常数已知的假设.两个数值仿真例子表明本方法的有效性.     

Guaranteed Cost Observer–Based Controls for a Class of Uncertain Neutral Time-Delay Systems

In this paper, guaranteed-cost observer-based controls for a class of uncertain neutral time-delay systems are considered. The asymptotic stabilization for the uncertain neutral systems is guaranteed with an observer-based feedback control. The linear matrix inequality (LMI) approach is used to design the observer-based feedback control system. Two classes of observer-based controls are proposed and their guaranteed costs are given. The control and observer gains are given from the LMI feasible solutions. A convex optimization problem with LMIs is formulated to design the optimal guaranteed-cost observer-based controls which minimize the guaranteed cost of the system considered. A numerical example is given to illustrate the results.The research reported here was supported by the National Science Council of Taiwan, ROC under Grant NSC 93-2213-E-214-020     

Robust Observer Design for a Class of Nonlinear Systems Using the System Internal Dynamics Structure

In this paper, an observer design strategy is presented for a class of nonlinear systems with structural uncertainty. The modern geometric approach is exploited to simplify the system structure. Then, based on the Lyapunov direct method, a robust observer is proposed using the system internal dynamics structure and the distribution of the uncertainty structure. The bound on the uncertainty, which is employed in the observer design, is allowed to be nonlinear and have a more general form. Simulation shows that the proposed approach is effective.     

Fuzzy Partial State Feedback Control of Discrete Nonlinear Systems with Unknown Time-Delay

A new discrete-time fuzzy partial state feedback control method for the nonlinear systems with unknown time-delay is proposed. Ma et al. proposed the design method of the fuzzy controller based on the fuzzy observer and Cao and Frank extend this result to be applicable to the case of the nonlinear systems with the time-delay. However, the time-delay is likely to be unknown in practical. In this paper, the sufficient condition for the asymptotic stability is derived with the assumption that the time-delay is unknown by applying Lyapunov–Krasovskii theorem and this condition is converted into the LMI problem.     

一类单时滞线性系统的模糊变结构控制

研究了一类单时滞线性系统变结构控制律的设计方法,基于模糊T-S模型把一类单时滞线性系统化为若干个时滞子系统,然后对时滞子系统设计变结构控制律,取全局控制作为系统的控制律,从而达到对单时滞线性系统系统进行控制的目的,给出了单时滞系统的滑模稳定的条件.仿真表明控制策略的有效性.     

不确定关联时滞大系统的自适应分散镇定控制

针对时变的满足一定匹配条件的不确定关联时滞大系统,利用自适应界化技术,给出了设计分散镇定控制器的自适应的方法.其特点是在假设中系统不确定项是有界的,但界是未知的,且在关联项存在时变时滞的情况下证明了闭环自适应系统的渐近稳定性.最后举例说明了该方法的有效性.     

一类不确定随机非线性系统的采样扩张状态观测器设计与观测误差的收敛性分析

实际系统中普遍存在各种干扰和不确定性因素,并且对控制系统的性能造成负面影响,因此设计对干扰和不确定性具有优异估计性能的观测器显得尤为重要.针对一类具有采样输出的不确定随机非线性系统,设计相应的采样扩张状态观测器用于在线估计不可量测状态和影响系统性能的随机总干扰.所估计的随机总干扰包含系统内部未建模动态、统计特性未知的外部有界噪声干扰以及不确定性因素的非线性耦合作用.在每个采样区间内,设计一个输出预估器用于估计实际输出,相应的输出估计值用于采样扩张状态观测器的设计.证明了所设计的采样扩张状态观测器对不可量测状态和随机总干扰的观测误差的均方收敛性.最后,一个具体仿真实例用于证实理论结果.     

基于输入输出线性化方法的一类非线性系统观测器设计

研究一类非线性系统的观测器设计问题.基于输入输出线性化方法提出了一类非线性系统的观测器设计.并且此非线性系统具有多输入多输出的特点,证明了在适当条件下,提出的观测器保证了观测误差渐近趋于零.仿真例表明了所得结果的有效性.     

一类非线性参数化时滞系统的鲁棒自适应衰减问题

研究了一类虚拟控制系数未知的非线性参数化时滞系统的自适应衰减问题.利用参数分离技术和backstepping设计方法构造自适应H∞控制器,使闭环系统在L2增益下从干扰输入到系统输出的影响任意小,并且无外部干扰时,系统内稳.     

非线性时滞系统的直接自适应模糊控制

针对一类具有摄动的严格反馈非线性时滞系统,基于后推设计方法,利用第一类模糊系统的逼近能力,提出了一种新的直接自适应控制方案。该方案避免了虚拟控制增益符号已知的假设。设计中引入连续鲁棒项对系统的摄动部分进行抑制。通过理论分析,证明了闭环系统是半全局一致终结有界的,跟踪误差收敛到一个小的残差集内。     

一类非线性系统的迭代学习控制

讨论一类非线性系统的迭代学习控制,系统的非线性动态对状态不快于多项式增长,而量测方程含有噪声.控制序列并非直接输给系统,而是先经过死区、预载及饱和等非线性函数.递推地给出了学习控制序列,并证明它的有界性及最优跟踪性.     

一类非线性微分代数系统的零动态

控制系统的零动态是系统一种内部动态品质,其行为与系统的许多性质相联系,如系统的稳定性,反馈镇定与输出跟踪等.针对一类非线性微分代数系统,提出了输出零子流形和零动态的概念.利用M-导数方法,探讨了此类系统的输出零子流形的性质,并给出了此类系统的输出零子流形和零动态算法,也讨论了该算法的一些性质.最后,给出一个例子说明如何利用系统的零动态来讨论系统的反馈稳定化问题.     

Observer Design for Systems with Diffusion

Finite-dimensional approximate observers are designed for one-dimensionaldiffusion processes described by a scalar partial differentialequation of the parabolic type. In the design, the observer'saccuracy can be traded off against its dynamical order and thenumber and type of the required measurements. The eigenvaluespectrum of the observation error dynamics is freely assignable.     

一类非线性哈密顿系统的周期解

结合Maslov指标理论,利用环绕定理证明了一类非线性哈密顿系统的周期解的存在性,而这类哈密顿系统所对应的作用泛函可能不满足Palais-Smale条件.     

一类非线性微分系统极限环的存在性

研究了非线性微分系统 (dx)/(dt)=p(y),(dy)/(dt)=-q(y)f(x)-g(x)极限环的存在性,获得了该系统包围多个奇点的极限环存在的两个充分条件,所获结果改进和推广了文[1,2,3]中的相应结果,并且指出了文[2,3,4,5]中的疏漏.     

Optimal Control of Nonlinear Time-Delay Systems with Persistent Disturbances

This paper is concerned with the optimal control of a class of nonlinear time-delay systems affected by external persistent disturbances. A feedforward and feedback optimal control (FFOC) law, consisting of analytical linear feedforward and feedback terms and the limit of a compensation sequence, is obtained by a successive approximation approach (SAA). In order to obtain a physically realizable feedforward control, a disturbance observer is introduced in the exosystem. Simulation results demonstrate the validity of the SAA and the robustness of the FFOC. Communicated by Q. C. Zhao This work was supported by the National Natural Science Foundation of China, Grant 60574023 and the Natural Science Foundation of Qingdao City of China, Grant 05-1-JC-94.     

一类单种群系统正周期解及其稳定性

考虑周期单种群模型 dxdt=xg( t,x)± p( t,x)的正周解及其稳定性 .证明了在一定条件下 ,系统存在全局吸引的正周期解 .给出了系统存在两个正周期解的充分条件 ,同时也给出了种群灭绝的条件 .这些结果用于 Logistic模型和 Odum模型 ,得到了被开发的周期 Logistic模型存在全局吸引的正周期解 ;被开发了的周期 Odum模型只存在两个正周期解 ,其中之一吸引初值大于一个定数的所有解 ,另一个周期解则是种群灭绝的分界线