2-D系统的干扰解耦

本文讨论了２－Ｄ系统的干扰解耦问题（ＤＤＰ），将１－Ｄ系统理论中的有关结果推广到了２－Ｄ线性离散常系数一般模型（２－ＤＧＭ），得到了问题可解的充分条件和必要条件以及相应的算法。这些结果较好地改进了现有结果。     

2－D系统的干扰解耦

本文讨论了２－Ｄ系统的干扰解耦问题（ＤＤＰ），将１－Ｄ系统理论中的有关结果推广到了２－Ｄ线性离散常系数一般模型（２－ＤＧＭ），得到了问题可解的充分条件和必要条件以及相应的算法．这些结果较好地改进了现有结果     

一般模型2－D系统的观测器设计理论

该文讨论了２－Ｄ一般模型（２ＤＧＭ）在标准边界条件下的渐近观测器的存在性条件及其设计问题．为此，首先将由Ｂｉｓｉａｃｃｏ等人在１９８５年发展起来的相应于对角边界条件的２－Ｄ渐近稳定性理论推广到了具有标准边界条件的２－Ｄ一般模型．在此基础上，借助于局部能控性概念，建立了一系列十分类似于１－Ｄ情形的观测器的存在条件，从这些存在条件出发也可以得到相应的观测器设计算法，最后还得出了２ＤＧＭ的分离性定理．     

2-D 奇异系统的实现

讨论了２－Ｄ奇异线性常系数一般模型（２－ＤＳＧＭ）在标准边界条件下的状态空间实现问题,提出了一种新的实现算法。     

2-D奇异系统的实现

讨论了２－Ｄ奇异线性常系数一般模型（２－ＤＳＧＭ）在标准边界条件下的状态空间实现问题,提出了一种新的实现算法。     

2－DRoesser模型的静态干扰解耦

本文讨论了２－ＤＲｏｅｓｓｅｒ模型［１］（ＲＭ）的静态干扰解耦问题［２］（简称为２－ＤＤＤＰ），即寻求２－Ｄ状态反馈使相应的闭环系统具有抗干扰的能力，得到了问题有解的充分条件和计算相应反馈阵的算法．     

Andrews型M-估计GM(1,1)模型在色谱保留值研究中的应用

提出了Ａｎｄｒｅｗｓ型Ｍ－估计ＧＭ（１，１）灰色模型，并用此模型对８种多环芳烃化合物容量因子与流动相组成之间的关系进行了关联。结果表明，该模型较常规ＧＭ（１，１）模型具有更好的抗干扰性能和受异常值影响小的优点，因此它可以取代常规ＧＭ（１，１）模型。     

一类非线性的自适应与L2增益混合控制

本文在文献（１，２）的基础上，对一类非线性系统提出了把自适应控制与Ｌ２－增益≤γ的控制相结合的控制思想，所导出的控制策略具有下面优点：１）整个系统仅需要与未知干扰个数相等的ｐ个估计器，克服了文献（１）中需要ｎｐ个估计器；２）扩充了文献（２）的结果，使未知扰动能够出现在控制通道，并能克服有界匹配外扰。计算机仿真结果表明控制方案是可行的。     

语音识别中广义模型及其算法收敛性分析

语音识别中,动态时间规整（Ｄｙｎａｍｉｃ Ｔｉｍｅ Ｗａｒｐｉｎｇ,简称ＤＴＷ）和隐马尔可夫模型（Ｈｉｄｄｅｎ Ｍａｒｋｏｖ Ｍｏｄｅｌ,简称 ＨＭＭ）是最有效的两种识别算法,并且 ＤＴＷ和 ＨＭＭ在本质上是一致的~［１］。根据 ＤＴＷ和 ＨＭＭ的本质联系和各自所对应的声学模型,在前期工作中建立了一种广义声学模型 ~［２］［３］（Ｇｅｎｅｒａｌ Ｍｏｄｅｌ,简称 ＧＭ）,并指出 ＤＴＷ和ＨＭＭ 只是 ＧＭ的特例,且 ＤＴＷ和 ＨＭＭ都可以转化为 ＧＭ。并在此基础上,首次将 Ｆｉｓｈｅｒ算法~［４］引进ＧＭ的学习算法,确保了ＧＭ状态分割的收敛性,并且这种分割在最小离差意义上是全局精确最优的。最后,从大数定理的角度出发,对 ＧＭ算法的收敛性进行了分析, 从理论上论证了该算法的依概率收敛性,并为实际应用中 ＧＭ算法的有效性提供了理论依据。     

含不可观变迁事件图的状态估计器设计

针对基于事件图建模的离散事件系统中存在的不可观变迁的状态估计问题,本文提出前向、后 向可观路径的概念对其进行了详细分析,并给出一种估计器设计方法,用来根据关联矩阵行变换进行不可观 变迁状态的估计．分析了估计器的观测代价,在此基础上给出了寻找最优估计器的方法,并利用最优估计器 计算不可观变迁的状态估计范围．最后通过实例证明了该方法能够有效地进行不可观变迁的状态估计．     

一般2-D线性常系数离散状态空间模型渐近稳定性的一类Lyapunov方法

本文给出了一般2-D线性常系数离散状态空间模型(2-D GM)的渐近稳定性的定义以及相应的判据,并借助于Lyapunov方程给出了2-D Roesser Model(2-D RM)渐近稳定条件,推广和简化了文献[1,7]的有关结论。最后通过行列式的一个简单性质,证明了关于2-DGM渐近稳定性的判定可以转化为一类特殊的2-D RM的相应问题,从而得到了判定2-D GM渐近稳定性的Lyapunov方法。     

二维随机FM-II系统的状态估计

This paper is concerned with state estimation of two-dimensional (2-D) discrete stochastic systems. First, 2-D discrete stochastic system model is established by extending system matrices of the well-known Fornasini-Marchesini's second model into stochastic matrices. Each element of these stochastic matrices is second-order weakly stationary white noise sequences. Secondly, a linear and unbiased full-order state estimation problem for 2-D discrete linear stochastic model is formulated. Two estimation problems considered are the designs for the mean-square bounded estimation error and for the mean-square stochastic version of the suboptimal H∞ estimator, respectively. Our results can be seen as extensions of the 2-D linear deterministic case. Finally, illustrative examples are provided.     

State estimation in discrete m-D systems

In this study observers and stochastic state estimation form- D systems are considered. The analysis uses a wave advance process model in transporting the existing 1-D results to them-D setting. A constrained minimization technique is used to compute the optimal state estimator which preserves quarter plane causality. The optimal state estimator, without the quarter plane causality constraint, is also determined. A comparison then yields insight into the tradeoff between quarter plane causality and estimator performance.     

一类存在数据丢失二维离散系统的H∞滤波

研宄输出测量数据丢失情况下二维线性离散系统的H_∞。滤波问题.首先,将数据丢失现象描述为随机伯努利序列,在此基础上建立二维系统状态估计误差的随机动态方程.其次,定义随机意义下的二维系统均方渐近稳定性和H_∞。性能,基于线性矩阵不等式给出误差系统满足均方渐近稳定和H_∞。性能的一个充分条件,该条件可以实现滤波器参数矩阵的设计.同时,研宄结果被进一步推广到不确定二维系统.最后,通过仿真示例验证了理论结果的有效性.     

Optimal control of 2-D systems

Two-dimensional (2-D) optimal control theory that parallels one-dimensional (1-D) optimal control is developed. A generalized performance measure suited to 2-D systems is introduced. The canonical equations associated with this performance measure and a general nonlinear model are obtained. The 2-D linear quadratic regulator problem is formulated, and its canonical equations are derived for the Roesser model. An earlier result by T. Kaczorek and J. Klamka (1986) for the solution of the minimum-energy problem with fixed-final local state is rederived using this approach. A new problem, minimum-energy with fixed-final-pass local states is formulated and solved, and a numerical example is given     

多传感器时滞系统信息融合最优Kalman滤波器

基于线性最小方差最优加权融合估计算法,对多传感器的离散线性状态时滞随机系统,给出了一种非增广分布式加权融合最优Kalman滤波器.推导了状态时滞系统任两个传感器子系统之间的滤波误差互协方差阵的计算公式.它与状态增广加权融合滤波器具有相同的精度.与每个传感器的局部滤波器相比,分布式融合滤波器具有更高的精度.与状态和观测增广最优滤波器相比,具有较小的精度.但避免了增广所带来的高维计算和大的空间存储,可减小计算负担.仿真例子验证了其有效性.     

Optimal continuous-time state estimation for linear finite and infinite-dimensional chemical process systems with state constraints

This work addresses optimal constrained state estimation problem for finite and infinite-dimensional chemical process systems. We consider cases when the prior information, in addition to the model parameters and the measurements, is available in the form of an inequality constraint with respect to the system's state. In the latest developments of the optimal state estimation theory, considerations of the state constraints have been often neglected since constraints do not fit easily in the structure of the optimal state estimator. Therefore, the issue of the state constraints being present needs to be addressed adequately, in particular, nonnegativity of concentration. Motivated by this, we developed a sequential, algorithmic optimal constrained state estimator for both finite and infinite-dimensional process systems commonly found in chemical process engineering (CSTR, tubular reactor). In this paper, we also designed an optimal constrained state estimator for a large class of dissipative infinite-dimensional systems which involve boundary actuation and point observation. Finally, illustrative examples of chemical process systems and proposed optimal state constrained estimation are presented.     

Stochastic control of processes having moving boundaries—An experimental study

A general state estimator requiring only discrete time measurements has been developed for nonlinear distributed parameter systems having moving boundaries. The state estimator has been combined with an optimal linear-quadratic feedback controller to provide a stochastic feedback control scheme for this class of problem. Both the state estimation and control algorithms were implemented in real time for a laboratory casting process and were found to perform well. The state estimation scheme would seem to hold the most interest for industrial applications.