求解约束优化问题的两个微分方程算法

本文给出求解具有等式约束和不等式约束的非线性优化问题的一阶信息和二阶信息的两个微分方程系统,问题的局部最优解是这两个微分方程系统的渐近稳定的平衡点,给出了这两个微分方程系统的Euler离散迭代格式并证明了它们的收敛性定理,用龙格库塔法分别求解两个微分方程系统．我们构造了搜索方向由两个微分系统计算,步长采用Armijo线搜索的算法分别求解这个约束最优化问题,在局部Lipschitz条件下基于二阶信息的微分方程系统的迭代方法具有二阶的收敛速度。我们给出的数值结果表明龙格库塔的微分方程算法具有较好的稳定性和更高的精确度,求解二阶信息的微分方程系统的方法具有更快的收敛速度．     

一个求解约束非线性优化问题的微分方程方法

本文构造的求解非线性优化问题的微分方程方法包括两个微分方程系统,第一个系统基于问题函数的一阶信息,第二个系统基于二阶信息．这两个系统具有性质:非线性优化问题的局部最优解是它们的渐近稳定的平衡点,并且初始点是可行点时,解轨迹都落于可行域中．我们证明了两个微分方程系统的离散迭代格式的收敛性定理和基于第二个系统的离散迭代格式的局部二次收敛性质．还给出了基于两个系统的离散迭代方法的数值算例,数值结果表明基于二阶信息的微分方程方法速度更快．     

一类偏积分微分方程二阶差分空间半离散格式的全局行为

偏积分微分方程产生于许多科学与工程领域,数值求解此类问题具有重要应用.本文给出了数值求解一类长时间偏积分微分方程的二阶差分空间半离散格式.借助于Laplace变换及Parseval等式,给出了全局稳定性的证明、误差估计及全局收敛性的结果.     

求解分布控制问题的预处理最小残量方法(英文)

通过分析Bai(Bai Z Z.Block preconditioners for elliptic PDE-constrained optimization problems.Computing,2011,91:379-395)给出的离散分布控制问题的块反对角预处理线性系统,提出了该问题的一个等价线性系统,并且运用带有预处理子的最小残量方法对该系统进行求解.理论分析和数值实验结果表明,所提出的预处理最小残量方法对于求解该类椭圆型偏微分方程约束最优分布控制问题非常有效,尤其当正则参数适当小的时候.     

三阶线性常微分方程Sinc方程组的结构预处理方法

三阶线性常微分方程在天文学和流体力学等学科的研究中有着广泛的应用.本文介绍求解三阶线性常微分方程由Sinc方法离散所得到的线性方程组的结构预处理方法.首先, 我们利用Sinc方法对三阶线性常微分方程进行离散,证明了离散解以指数阶收敛到原问题的精确解.针对离散后线性方程组的系数矩阵的特殊结构, 提出了结构化的带状预处理子,并证明了预处理矩阵的特征值位于复平面上的一个矩形区域之内.然后, 我们引入新的变量将三阶线性常微分方程等价地转化为由两个二阶线性常微分方程构成的常微分方程组, 并利用Sinc方法对降阶后的常微分方程组进行离散.离散后线性方程组的系数矩阵是分块2×2的, 且每一块都是Toeplitz矩阵与对角矩阵的组合.为了利用Krylov子空间方法有效地求解离散后的线性方程组,我们给出了块对角预处理子, 并分析了预处理矩阵的性质.最后, 我们对降阶后二阶线性常微分方程组进行了一些比较研究.数值结果证实了Sinc方法能够有效地求解三阶线性常微分方程.     

一类带弱奇异核偏积分微分方程空间半离散的Legendre-Galerkin谱方法

本文给出了数值求解一类偏积分微分方程的空间半离散Legendre-Galerkin谱方法;证明了解的稳定性及收敛性。     

线性常微分方程的全局误差估计和优化求解方法

线性常微分方程初值问题求解在许多应用中起着重要作用.目前,已存在很多的数值方法和求解器用于计算离散网格点上的近似解,但很少有对全局误差(global error)进行估计和优化的方法.本文首先通过将离散数值解插值成为可微函数用来定义方程的残差;再给出残差与近似解的关系定理并推导出全局误差的上界;然后以最小化残差的二范数为目标将方程求解问题转化为优化求解问题;最后通过分析导出矩阵的结构,提出利用共轭梯度法对其进行求解.之后将该方法应用于滤波电路和汽车悬架系统等实际问题.实验分析表明,本文估计方法对线性常微分方程的初值问题的全局误差具有比较好的估计效果,优化求解方法能够在不增加网格点的情形下求解出线性常微分方程在插值解空间中的全局最优解.     

Richardson外推及其推广

本文分别针对微分方程数值求解和数值积分两类问题,讨论了一种常见的加速收敛方法-R ichardson外推及其推广,同时给出了一定的理论分析,对之进行了对比数值实验,验证了理论的正确性.     

一类变延迟微分方程谱方法的收敛性

本文主要应用谱方法求解一类线性变系数变延迟微分方程,构造相应的数值方法,证明其收敛性,并给出两个具有代表性的数值算例.这些结果表明应用谱方法求解延迟微分方程可以获得谱收敛与谱精度的计算效果.     

多约束二阶非线性常微分方程拐点的数值求法

提出了一种求解多约束二阶非线性常微分方程拐点的数值解法,并以具有平行反应的化学放热系统为例,给出了一个具体算例.     

New numerical methods and some applied aspects of the p-regularity theory

The theory of p-regularity is applied to optimization problems and to singular ordinary differential equations (ODE). The special variant of the method of the modified Lagrangian function proposed by Yu.G. Evtushenko for constrained optimization problems with inequality constraints is justified on the basis of the 2-factor transformation. An implicit function theorem is given for the singular case. This theorem is used to show the existence of solutions to a boundary value problem for a nonlinear differential equation in the resonance case. New numerical methods are proposed including the p-factor method for solving ODEs with a small parameter.     

Two differential equation systems for equality-constrained optimization

This paper presents two differential systems, involving first and second order derivatives of problem functions, respectively, for solving equality-constrained optimization problems. Local minimizers to the optimization problems are proved to be asymptotically stable equilibrium points of the two differential systems. First, the Euler discrete schemes with constant stepsizes for the two differential systems are presented and their convergence theorems are demonstrated. Second, we construct algorithms in which directions are computed by these two systems and the stepsizes are generated by Armijo line search to solve the original equality-constrained optimization problem. The constructed algorithms and the Runge–Kutta method are employed to solve the Euler discrete schemes and the differential equation systems, respectively. We prove that the discrete scheme based on the differential equation system with the second order information has the locally quadratic convergence rate under the local Lipschitz condition. The numerical results given here show that Runge–Kutta method has better stability and higher precision and the numerical method based on the differential equation system with the second information is faster than the other one.     

Modified block-approximate factorization strategies

Summary Two variants of modified incomplete block-matrix factorization with additive correction are proposed for the iterative solution of large linear systems of equations. Both rigorous theoretical support and numerical evidence are given displaying their efficiency when applied to discrete second order partial differential equations (PDEs), even in the case of quasi-singular problems.Research supported by the A.B.O.S. (A.G.C.D.) under project 11, within the co-operation between Belgium and Zaire     

Trajectory-following algorithms for min-max optimization problems

In this paper, we consider a class of nonlinear minimum-maximum optimization problems subject to boundedness constraints on the decision vectors. Three algorithms are developed for finding the min-max point using the concept of solving an associated dynamical system. In the first and third algorithms, solutions are obtained by solving systems of differential equations. The second algorithm is a discrete version of the first algorithm. The trajectories generated by the first and second algorithms may move inside or on the boundary of the constraint set, while the third algorithm ensures that any trajectory that begins inside the constraint region remains in its interior. Sufficient conditions for global convergence of the two algorithms are also established. For illustration, four numerical examples are solved.This work was partially supported by a research grant from the Australian Research Committee.     

Riemann‐Hilbert problem and matrix discrete Painlevé II systems

Matrix Szeg? biorthogonal polynomials for quasi‐definite matrices of Hölder continuous weights are studied. A Riemann‐Hilbert problem is uniquely solved in terms of the matrix Szeg? polynomials and its Cauchy transforms. The Riemann‐Hilbert problem is given as an appropriate framework for the discussion of the Szeg? matrix and the associated Szeg? recursion relations for the matrix orthogonal polynomials and its Cauchy transforms. Pearson‐type differential systems characterizing the matrix of weights are studied. These are linear systems of ordinary differential equations that are required to have trivial monodromy. Linear ordinary differential equations for the matrix Szeg? polynomials and its Cauchy transforms are derived. It is shown how these Pearson systems lead to nonlinear difference equations for the Verblunsky matrices and two examples, of Fuchsian and non‐Fuchsian type, are considered. For both cases, a new matrix version of the discrete Painlevé II equation for the Verblunsky matrices is found. Reductions of these matrix discrete Painlevé II systems presenting locality are discussed.     

On asymptotic equivalence of linear and quasilinear difference equations

The asymptotic equivalence of systems of difference equations of linear and quasilinear type is investigated. The first result on the asymptotic equivalence of linear systems is a discrete analog of an improved version of the Levinson's well-known theorem on asymptotic equivalence of linear differential equations, while the second one providing conditions for asymptotic equivalence of linear and quasilinear systems is related to that of Yakubovich in differential equations case.     

Simple models for biological control of crop pests and their application

In this work, we construct simple models in terms of differential equations for the dynamics of pest populations and their management using biological pest control. For the first model used, the effect of the biological control is modelled by a function of repeated infinite impulses. And, our second model uses a periodic function proportional to the population to model the effect of biological control. In both cases, we present analytical solutions and derive a discrete version of them. Moreover, convergence conditions are given for periodic solutions. Finally, an application of such models is described for diamondback moth in a plot of broccoli to be controlled by the application of biological pesticides and beneficial parasitoids.     

A fully Sinc-Galerkin method for Euler–Bernoulli beam models

A fully Sinc-Galerkin method in both space and time is presented for fourth-order time-dependent partial differential equations with fixed and cantilever boundary conditions. The sine discretizations for the second-order temporal problem and the fourth-order spatial problems are presented. Alternate formulations for variable parameter fourth-order problems are given, which prove to be especially useful when applying the forward techniques of this article to parameter recovery problems. The discrete system that corresponds to the time-dependent partial differential equations of interest are then formulated. Computational issues are discussed and an accurate and efficient algorithm for solving the resulting matrix system is outlined. Numerical results that highlight the method are given for problems with both analytic and singular solutions as well as fixed and cantilever boundary conditions.