一种基于方差约束H∞鲁棒滤波新算法

基于方差约束,研究一类不确定线性定常随机离散系统的H∞滤波问题。提出了一种鲁棒滤波的新算法,该算法克服构造对角矩阵约束性较强,缺乏一般性的缺点,给出了基于线性矩阵不等式(LMI)的同时满足方差和H∞约束的鲁棒滤波器,且在自由参数允许的前提下,与同类算法相比,其滤波增益较小,滤波误差协方差也较小。文中给出的数值算例验证了本算法的有效性。     

Robust Stability of Networked Load Frequency Control Systems with Time-Varying Delays

In deregulated power system scenario coupled with smart grid technology, for networked load frequency control (LFC), an open communication structure, owing to low cost and flexibility, is preferred over dedicated networks in the feedback control loop for transmitting/receiving the data between the geographically displaced power system and the control center. In such a control scheme, closing the feedback loop through an open communication channel, in turn, introduces two additive time-varying delays of dissimilar characteristics in the feedback path. These delays degrade the performance of the closed-loop system, and exert a destabilizing effect on the overall system. In this paper, using Lyapunov-Krasovskii functional approach, a less conservative stability criterion is presented to ascertain delay-dependent stability of such network-controlled LFC systems with two additive time-varying delays in the feedback path in the presence of uncertain load disturbance conditions. Unlike the existing results, which are derived by combining these two delays into one, the proposed result considers the two delays as separate entities thereby imparting more generalization into the stability analysis. The effect of unknown exogenous load disturbance is incorporated by mathematically modeling them as a bounded non-linear time-varying function of current and delayed state vectors.     

The great asymmetric divide: An empirical investigation of the link between indigenous and non‐indigenous economic systems in Northern Australia

This empirical study explores financial links between indigenous and non‐indigenous economic systems in a remote river catchment in Northern Australia (the Mitchell). It finds evidence of a profound and asymmetric ‘disconnect’ between these economies: an exogenous increase in indigenous incomes raises the incomes of non‐indigenous people, but the reverse is not true. Evidently, those seeking to improve the incomes of indigenous people in Northern Australia cannot simply seek to (i) increase payments to indigenous people, or (ii) expand the non‐indigenous sector hoping that some benefits will ‘trickle down’. Instead, structural change is required.     

Design of robust reliable control for T-S fuzzy Markovian jumping delayed neutral type neural networks with probabilistic actuator faults and leakage delays: An event-triggered communication scheme

This study examines the problem of robust reliable control for Takagi-Sugeno (T-S) fuzzy Markovian jumping delayed neural networks with probabilistic actuator faults and leakage terms. An event-triggered communication scheme. First, the randomly occurring actuator faults and their failures rates are governed by two sets of unrelated random variables satisfying certain probabilistic failures of every actuator, new type of distribution based event triggered fault model is proposed, which utilize the effect of transmission delay. Second, Takagi-Sugeno (T-S) fuzzy model is adopted for the neural networks and the randomness of actuators failures is modeled in a Markov jump model framework. Third, to guarantee the considered closed-loop system is exponential mean square stable with a prescribed reliable control performance, a Markov jump event-triggered scheme is designed in this paper, which is the main purpose of our study. Fourth, by constructing appropriate Lyapunov-Krasovskii functional, employing Newton-Leibniz formulation and integral inequalities, several delay-dependent criteria for the solvability of the addressed problem are derived. The obtained stability criteria are stated in terms of linear matrix inequalities (LMIs), which can be checked numerically using the effective LMI toolbox in MATLAB. Finally, numerical examples are given to illustrate the effectiveness and reduced conservatism of the proposed results over the existing ones, among them one example was supported by real-life application of the benchmark problem.     

基于临界Galton -Watson过程的随机游动的大偏差

针对一族独立同分布的随机变量{X_k}的和S_{Z<sup>^</sup>n}=^Z_<sup>n_k=1X_k(Z_n为临界Galton-Watson过程的第n代个体数),利用随机游动和概率论的知识研究了R_n:=S_{Z<sup>^</sup>n}/Z_n的渐近性质以及在{Z_n>0}条件下的S_{Z<sup>^</sup>n}的大偏差.研究结果表明, R_n的规范偏差概率有非退化的极限,并且其大偏差规范化后收敛到正常数.     

线性跟踪系统Lyapunov函数性质及其应用

研究了线性跟踪系统Lyapunov函数性质,并基于该性质进行控制系统设计研究,给出控制器结构和Lyapunov函数取值估计;首先,以经典三回路为基础建立线性闭环跟踪系统微分方程;在完成系统数学模型的基础上,给出线性跟踪系统Lyapunov函数对时间的一阶导数为非负,二阶导数为非正,即线性跟踪系统Lyapunov函数最大值不是在系统的初始状态而是稳定状态这一论点;进而基于该Lyapunov函数性质对控制系统进行设计和Lyapunov函数取值进行研究;最后通过仿真验证,仿真结果表明所提出的线性跟踪系统Lyapunov函数性质、控制器设计和Lyapunov函数取值估计是正确的。     

Generalized eigenvalue minimization for uncertain first-order plus time-delay processes

This paper shows how to apply generalized eigenvalue minimization to processes that can be described by a first-order plus time-delay model with uncertain gain, time constant and delay. An algorithm to transform the uncertain first-order plus time delay model into a state-space model with uncertainty polyhedron is firstly described. The accuracy of the transformation is studied using numerical examples. Then, the uncertainty polyhedron is rewritten as a linear-matrix-inequality constraint and generalized eigenvalue minimization is adopted to calculate a feedback control law. Case studies show that even if uncertainties associated with the first-order plus time delay model are significant, a stable feedback control law can be found. The proposed control is tested by comparing with a robust internal model control. It is also tested by applying it to the temperature control of air-handing units.