一类离散时间代数Riccati矩阵方程对称解的双迭代算法

利用逆矩阵的Neumann级数形式,将在线性二次优化问题中遇到的含未知矩阵之逆的离散时间代数Riccati矩阵方程(DTARME)转化为高次多项式矩阵方程,然后采用牛顿算法求高次多项式矩阵方程的对称解,并采用修正共轭梯度法求由牛顿算法每一步迭代计算导出的线性矩阵方程的对称解或者对称最小二乘解,建立求DTARME的对称解的双迭代算法。双迭代算法仅要求DTARME有对称解,不要求它的对称解唯一,也不对它的系数矩阵做附加限定。数值算例表明双迭代算法是有效的。     

Riccati方程子矩阵约束对称解的非精确Newton-MCG算法

采用修正共轭梯度法(MCG算法)求由Newton算法每一步迭代计算导出的线性矩阵方程的近似子矩阵约束(SMC)对称解或者近似SMC对称最小二乘解,建立求离散时间代数Riccati矩阵方程SMC对称解的非精确Newton-MCG算法.该算法仅要求Riccati矩阵方程有SMC对称解,不要求它的SMC对称解唯一,也不要求导出的线性矩阵方程有相应的SMC对称解.数值算例表明,非精确Newton-MCG算法是有效的.     

求矩阵方程AXB=C的双对称最小二乘解的迭代算法

基于求解线性代数方程组的共轭梯度法的思想,通过特殊的变形与近似处理,建立了求矩阵方程AXB=C的双对称最小二乘解的迭代算法,并证明了迭代算法的收敛性.不考虑舍入误差时,迭代算法能够在有限步计算之后得到矩阵方程的双对称最小二乘解;选取特殊的初始矩阵时,还能够求得矩阵方程的极小范数双对称最小二乘解.同时,也能够给出指定矩阵的最佳逼近双对称矩阵.算例表明,迭代算法是有效的.     

多矩阵变量线性矩阵方程的广义自反解的迭代算法

基于求线性矩阵方程约束解的修正共轭梯度法的思想方法,通过修改某些矩阵的结构,建立了求特殊类型的多矩阵变量线性矩阵方程的广义自反解的迭代算法,证明了迭代算法的收敛性,解决了给定矩阵在该矩阵方程的广义自反解集合中的最佳逼近计算问题.当矩阵方程相容时,该算法可以在有限步计算后得到其一组广义自反解;选取特殊的初始矩阵,能够求得其极小范数广义自反解.数值算例表明,迭代算法是有效的.     

关于一般耦合矩阵方程的迭代对称解

本文构造了一个有效的迭代方法(CGL)去求解一般耦合矩阵方程的对称解.若一般耦合矩阵方程关于对称解相容,则对于任意给定的初始对称矩阵组,利用所构造的迭代算法,都能在有限步迭代出所求问题的一组对称解,若选用一些特殊的初值,则可获得矩阵方程的极小范数对称解.最后的数值例子表明了所给算法的有效性.     

H∞滤波问题数值求解的精细积分算法

有限时间H∞滤波的Riccati方程和滤波方程分别为非线性矩阵微分方程和线性变系 数微分方程,而且Riccati微分方程解的存在性还依赖于参数 γ-2,因此求这些方程的数值解一 般比较困难.按照结构力学与最优控制的模拟关系,Riccati方程解存在的临界参数 γ-2cr对应于 广义Rayleigh商的一阶本征值.因此可以用精细积分法结合扩展的Wittrick-Williams(W-W) 算法计算 γ-2cr .并求解Ricclati方程,滤波微分方程的解也可以由精细积分法计算.     

求多变量线性矩阵方程组自反解的迭代算法

利用矩阵分解的方法求多变量线性矩阵方程组的自反解是很困难的.本文建立了一种迭代方法来解决这个问题,利用此迭代方法可以判断多变量线性矩阵方程组的可解性,且当矩阵方程组相容时,可以在有限步迭代后得到其自反解.选取特殊的初始矩阵时,能够求得矩阵方程组的极小范数自反解.进一步,通过求新的线性矩阵方程组的极小范数自反解,能够求得给定矩阵的最佳逼近矩阵.数值算例表明,迭代算法是有效的.     

线性矩阵方程异类约束最小二乘解的迭代算法

多矩阵变量线性矩阵方程(LME)约束解的计算问题在参数识别、结构设计、振动理论、自动控制理论等领域都有广泛应用。本文借鉴求线性矩阵方程(LME)同类约束最小二乘解的迭代算法,通过构造等价的线性矩阵方程组,建立了求多矩阵变量LME的一种异类约束最小二乘解的迭代算法,并证明了该算法的收敛性。在不考虑舍入误差的情况下,利用该算法不仅可在有限步计算后得到LME的一组异类约束最小二乘解,而且选取特殊初始矩阵时,可求得LME的极小范数异类约束最小二乘解。另外,还可求得指定矩阵在该LME的异类约束最小二乘解集合中的最佳逼近解。算例表明,该算法是有效的。     

基于策略迭代的连续时间系统的随机线性二次最优控制

针对模型参数部分未知的随机线性连续时间系统, 通过策略迭代算法求解无限时间随机线性二次(LQ) 最优控制问题. 求解随机LQ最优控制问题等价于求随机代数Riccati 方程(SARE) 的解. 首先利用伊藤公式将随机微分方程转化为确定性方程, 通过策略迭代算法给出SARE 的解序列; 然后证明SARE 的解序列收敛到SARE 的解, 而且在迭代过程中系统是均方可镇定的; 最后通过仿真例子表明策略迭代算法的可行性.

     

摄动离散矩阵Lyapunov方程解的估计

研究摄动离散矩阵Lyapunov方程解的估计问题,利用矩阵运算性质及Lyapunov稳定性理论,给出在结构不确定性假设下方程解的存在条件及解的上下界估计,估计结果由一个线性矩阵不等式(LMI)和两个矩阵代数Riccati方程确定．针对几种不确定性假设,进一步给出矩阵代数Riccati方程的具体形式．最后通过一个算例说明了所得结果的有效性．     

Numerical methods for the minimal non‐negative solution of the non‐symmetric coupled algebraic Riccati equation

In this paper, base on the theory of the non‐symmetric algebraic Riccati equation and the coupled Riccati equation, we give the general form of the non‐symmetric coupled algebraic Riccati equation (NCARE). In order to effectively solve the minimal non‐negative solution of the NCARE, two numerical iteration methods are improved: inexact Newton method (INewton) and alternate linear implicit method (ALI). Further, we give the convergence theory of the two iteration methods. Finally, we offer numerical examples to show the effectiveness of the derived methods.     

Criterion for the convergence of the solution of the Riccati differential equation

The optimal control problem for a linear system with a quadratic cost function leads to the matrix Riccati differential equation. The convergence of the solution of this equation for increasing time interval is investigated as a function of the final state penalty matrix. A necessary and sufficient condition for convergence is derived for stabilizable systems, even if the output in the cost function is not detectable. An algorithm is developed to determine the limiting value of the solution, which is one of the symmetric positive semidefinite solutions of the algebraic Riccati equation. Examples for convergence and nonconvergence are given. A discussion is also included of the convergence properties of the solution of the Riccati differential equation to any real symmetric (not necessarily positive semidefinite) solution of the algebraic Riccati equation.     

线性时滞系统的无源控制

研究一类线性时滞系统通过线性无记忆状态反馈控制律的无源控制问题。通过某个Ｒｉｃｃａｔｉ矩阵方程对称正定解的存在性,给出了使得闭环系统严格无源的控制器存在条件。进而,利用这个方程的正定解给出了无源化控制器的一个构造方法。     

基于GPU-CUDA的共轭斜量法实现及性能对比

偏微分方程数值解法(包括有限差分法、有限元法)以及大量的数学物理方程数值解法最终都会演变成求解大型线性方程组。因此,探讨快速、稳定、精确的大型线性方程组解法一直是数值计算领域不断深入研究的课题且具有特别重要的意义。在迭代法中,共轭斜量法(又称共轭梯度法)被公认为最好的方法之一。但是,该方法最大缺点是仅适用于线性方程组系数矩阵为对称正定矩阵的情况,而且常规的CPU算法实现非常耗时。为此,通过将线性方程组系数矩阵作转换成对称矩阵后实施基于GPU-CUDA的快速共轭斜量法来解决一般性大型线性方程组的求解问题。试验结果表明:在求解效率方面,基于GPU-CUDA的共轭斜量法运行效率高,当线性方程组阶数超过3000时,其加速比将超过14;在解的精确性与求解过程的稳定性方面,与高斯列主元消去法相当。基于GPU-CUDA的快速共轭斜量法是求解一般性大型线性方程组快速而非常有效的方法。     

Robust stabilization of uncertain systems with time-varyingmultistate delay

The authors deal with the problem of stabilizing a class of uncertain linear systems with time-varying multi-state delay and subject to norm-bounded parameter uncertainty via memoryless linear state feedback. Some sufficient conditions for the robust stabilizability are derived fur this class of uncertain systems. If there exists a positive-definite symmetric solution satisfying the algebraic Riccati equation (or inequality), a suitable memoryless state feedback law can be derived also. Moreover, all such parametric algebraic Riccati inequalities have been transformed into some linear matrix inequality problems, so there is no tuning of the parameters to gain a stabilizing solution     

Continuation methods for solving modified discrete-time algebraicRiccati equations

It is well known that the continuation methods have been successfully applied to solve polynomial systems and fixed point problems, etc. In this paper, we consider a discrete-time algebraic Riccati equation with an admissible, low rank, and symmetric perturbation. Our attention is directed primarily to this modified discrete-time algebraic Riccati equation and the numerical method for its solution based on proceeding along the continuation path     

Robust stabilization of linear systems with norm-bounded time-varying uncertainty

In this paper, robust stabilization of a class of linear systems with norm-bounded time-varying uncertainties is considered. It is shown that for this class of uncertain systems quadratic stabilizability via linear control is equivalent to the existence of a positive definite symmetric matrix solution to a (parameter-dependent) Riccati equation. Also, a construction for the stabilizing feedback law is given in terms of the solution to the Riccati equation.     

Weighted steepest descent method for solving matrix equations

This paper describes iterative methods for solving the general linear matrix equation including the well-known Lyapunov matrix equation, Sylvester matrix equation and some related matrix equations encountered in control system theory, as special cases. We develop the methods from the optimization point of view in the sense that the iterative algorithms are constructed to solve some optimization problems whose solutions are closely related to the unique solution to the linear matrix equation. Actually, two optimization problems are considered and, therefore, two iterative algorithms are proposed to solve the linear matrix equation. To solve the two optimization problems, the steepest descent method is adopted. By means of the so-called weighted inner product that is defined and studied in this paper, the convergence properties of the algorithms are analysed. It is shown that the algorithms converge at least linearly for arbitrary initial conditions. The proposed approaches are expected to be numerically reliable as only matrix manipulation is required. Numerical examples show the effectiveness of the proposed algorithms.