Development of a family of explicit algorithms for structural dynamics with unconditional stability

A new family of explicit integration algorithms is developed based on discrete control theory for solving the dynamic equations of motion. The proposed algorithms are explicit for both displacement and velocity and require no factorisation of the damping matrix and the stiffness matrix. Therefore, for a system with nonlinear damping and stiffness, the proposed algorithms are more efficient than the common explicit algorithms that provide only explicit displacement. Accuracy and stability properties of the proposed algorithms are analysed theoretically and verified numerically. Certain subfamilies are found to be unconditionally stable for any system state (linear elastic, stiffness softening or stiffness hardening) that may occur in earthquake engineering of a practical structure. With dual explicit expression and excellent stability property, the proposed family of algorithms can potentially solve complicated nonlinear dynamic problems.     

TWO-GRID ERROR ESTIMATES FOR THE STREAM FUNCTION FORM OF NAVIER-STOKES EQUATIONS

ＩｎｔｒｏｄｕｃｔｉｏｎＷｅｃｏｎｓｉｄｅｒｔｗｏ＿ｇｒｉｄｍｅｔｈｏｄｆｏｒｔｈｅｓｔｒｅａｍｆｕｎｃｔｉｏｎｆｏｒｍｏｆｔｈｅｓｔａｔｉｏｎａｒｙＮａｖｉｅｒ＿Ｓｔｏｋｅｓｅｑｕａｔｉｏｎｓ.Ｔｈｅａｄｖａｎｔａｇｅｓｏｆｔｈｅｓｔｒｅａｍｆｕｎｃｔｉｏｎｆｏｒｍａｒｅｔｈａｔｔｈｅｉｎｃｏｍｐｒｅｓｓｉｂｉｌｉｔｙｃｏｎｄｉｔｉｏｎｉｓｓａｔｉｓｆｉｅｄａｕｔｏｍａｔｉｃａｌｌｙａｎｄｔｈｅｐｒｅｓｓｕｒｅｉｓｎｏｔｐｒｅｓｅｎｔｉｎｔｈｅｗｅａｋｆｏｒｍ .Ｔｈｅｍｅｔｈｏｄｉｓｂａｓｅｄ…     

非半单分叉的Normal Form计算

在文［１］基础上，提出一种仅知道派生线性系统零实部特征值时求解非线性系统非半单分叉ＮｏｒｍａｌＦｏｒｍ的方法．通过适当的分类，将要求解的线性代数方程组分为若干相互独立的方程组．将所求系数向量按字典序列排列后，各独立方程组的系数矩阵是上三角矩阵．在非共振情形，各系数向量可按反字典序列递推求出．在共振情形，根据文中的二个定理，巧妙地由一简单的常数矩阵的最大秩子矩阵，定位其系数矩阵的满秩子矩阵，解决了这类方程组的降维简化．通过消元法，把简化后的方程化成类似于半单分叉ＮｏｒｍａｌＦｏｒｍ求解过程中方程的形式，其解法也类似．该方法非常易于在计算机代数软件平台上程序化．     

分数阶微分方程的Haar小波算法研究

在BPFs的Caputo分数阶微分算子矩阵的基础上,建立了Haar小波的分数阶微分算子矩阵,提出了一种有效的求解分数阶微分方程的Haar小波数值方法,并将该方法应用于线性和非线性分数阶常微分方程求解中.数值算例表明,该算法简单,数值精确度高,是一种高效的数值求解方法.     

Globally convergent homotopy algorithms for nonlinear systems of equations

Probability-one homotopy methods are a class of algorithms for solving nonlinear systems of equations that are accurate, robust, and converge from an arbitrary starting point almost surely. These new globally convergent homotopy techniques have been successfully applied to solve Brouwer fixed point problems, polynomial systems of equations, constrained and unconstrained optimization problems, discretizations of nonlinear two-point boundary value problems based on shooting, finite differences, collocation, and finite elements, and finite difference, collocation, and Galerkin approximations to nonlinear partial differential equations. This paper introduces, in a tutorial fashion, the theory of globally convergent homotopy algorithms, deseribes some computer algorithms and mathematical software, and presents several nontrivial engineering applications.This work was supported in part by DOE Grant DE-FG05-88ER25068, NASA Grant NAG-1-1079, and AFOSR Grant 89-0497.     

求解非线性方程组的混合遗传算法

非线性方程组的求解是数值计算领域中最困难的问题。大多数的数值求解算法例如牛顿法的收敛性和性能特征在很大程度上依赖于初始点。但是对于很多非线性方程组,选择好的初始点是一件非常困难的事情。本文结合遗传算法和经典算法的优点,提出了一种用于求解非线性方程组的混合遗传算法。该混合算法充分发挥了遗传算法的群体搜索和全局收敛性,有效地克服了经典算法的初始点敏感问题;同时在遗传算法中引入经典算法(Powell法、拟牛顿迭代法)作局部搜索,克服了遗传算法收敛速度慢和精度差的缺点。选择了几个典型非线性方程组,从收敛可靠性、计算成本和适用性等指标分析对比了不同算法。计算结果表明所设计的混合遗传算法有着可靠的收敛性和较高的收敛速度和精度,是求解非线性方程组的一种成功算法。     

小波方法及其非线性力学问题应用分析

小波分析是近几十年来发展起来的重要数学分支,被誉为“数学显微镜”,其独具的多分辨分析和大量可供选择的,可兼具正交性、紧支性、对称性、低通滤波、线性相位及插值性等优良数学品质的小波基函数为强非线性微分方程的数值求解带来了新的契机。自上世纪90年代以来,诸如小波伽辽金法、小波配点法、小波有限单元法和小波边界单元法等数值方法被先后构建出来并成功应用于各类力学问题的定量研究之中。本文从小波提出的历史背景及作为其理论基础的多分辨分析出发,对现有基于小波理论的各类数值方法进行梳理,总结各自的优点、缺点和下一步可能的发展方向,为未来基于小波理论的定量分析方法的发展及其在复杂非线性力学问题中的应用研究提供参考。     

一种改进的2n因子乘积形式的逆矩阵求解算法

复杂结构的有限元分析需要耗费巨大的内存和计算机时间。在此过程中，有限元线性方程组的求解时间占很大比例，因此，发展高效的线性方程组求解算法是提高有限元分析效率的关键，针对矩阵求逆问题，该文将有限元方程组的系数矩阵视为等分块矩阵，并基于等分块矩阵的概念推广了传统的逆矩阵的n因子和2n因子乘积形式。推广后得到的这组乘积形式在适当条件下又分别可以退化到传统的逆矩阵的n因子2n因子乘积形式，应用基于推广后得到的这组乘积形式的求逆算法来求解板材冲压成形有限元数值模拟中的大型有限元线性方程组，结果表明，该算法可以显著地提高大型有限元线性方程组的求解效率。     

Nonlinear axisymmetric deformations of an elastic tube under external pressure

The problem of the finite axisymmetric deformation of a thick-walled circular cylindrical elastic tube subject to pressure on its external lateral boundaries and zero displacement on its ends is formulated for an incompressible isotropic neo-Hookean material. The formulation is fully nonlinear and can accommodate large strains and large displacements. The governing system of nonlinear partial differential equations is derived and then solved numerically using the C++ based object-oriented finite element library Libmesh. The weighted residual-Galerkin method and the Newton-Krylov nonlinear solver are adopted for solving the governing equations. Since the nonlinear problem is highly sensitive to small changes in the numerical scheme, convergence was obtained only when the analytical Jacobian matrix was used. A Lagrangian mesh is used to discretize the governing partial differential equations. Results are presented for different parameters, such as wall thickness and aspect ratio, and comparison is made with the corresponding linear elasticity formulation of the problem, the results of which agree with those of the nonlinear formulation only for small external pressure. Not surprisingly, the nonlinear results depart significantly from the linear ones for larger values of the pressure and when the strains in the tube wall become large. Typical nonlinear characteristics exhibited are the “corner bulging” of short tubes, and multiple modes of deformation for longer tubes.     

THE THEORETICAL COST OF SEQUENTIAL AND PARALLEL ALGORITHMS FOR SOLVING LINEAR SYSTEMS OF EQUATIONS

ＴＨＥ　ＴＨＥＯＲＥＴＩＣＡＬ　ＣＯＳＴ　ＯＦ　ＳＥＱＵＥＮＴＩＡＬ　ＡＮＤ　ＰＡＲＡＬＬＥＬ　ＡＬＧＯＲＩＴＨＭＳ　ＦＯＲ　ＳＯＬＶＩＮＧ　ＬＩＮＥＡＲ　ＳＹＳＴＥＭＳ　ＯＦ　ＥＱＵＡＴＩＯＮＳＳａｌｍａｎＨ．Ａｂｂａｓ（ＲｅｃｅｉｖｅｄＭａｙ２４...     

一种求解局部非线性结构稳态响应的方法

为了降低求解局部非线性结构稳态响应的计算量,基于子结构和阻抗缩聚提出了一种用于求解局部非线性结构稳态响应的计算方法.将局部非线性结构分解为线性子结构和非线性子结构,利用谐波平衡构造各个子结构的阻抗方程,对线性子结构进行缩聚,将局部非线性动力学方程转化为求解一组非线性代数方程组问题,通过迭代求解非线性代数方程组,求解系统的稳态响应.     

松弛方法在计算凝聚炸药爆轰问题中的应用

利用松弛近似,将非线性的凝聚炸药爆轰控制方程转化为线性的松弛方程组,并采用五阶WENO格式和五阶线性多步显隐格式对线性松弛方程组进行空间方向和时间方向的离散,由此建立具有高精度和高分辨率性质的计算凝聚炸药爆轰的松弛方法。建立的松弛方法可以避免求解Riemann问题及计算非线性通量的Jacobi矩阵,同时无需分裂处理反应源项。通过对凝聚炸药的平面一维定常爆轰波结构及球面一维聚心、散心爆轰起爆和传播过程的数值模拟,验证了所建立的松弛方法能够很好地计算凝聚炸药爆轰问题。     

Variational methods of solving dynamic problems for fluid-containing bodies

A variational approach to solving linear and nonlinear problems for a body with cavities partially filled with a perfect incompressible fluid is enunciated. The approach applies a nonclassical variational principle to describe the spatial motion of a finite fluid with a free surface and the classical variational principle, which is widely used in rigid body dynamics. These principles are used to formulate variational problems that are the basis of direct methods of solving nonlinear and linear dynamic problems for body-fluid systems. The approach allows us to derive an infinite system of nonlinear ordinary differential equations describing the joint motion of the rigid body and fluid and to develop an algorithm for determining the hydrodynamic coefficients. Linearized differential equations of motion of the mechanical system are presented and approximate methods are given to solve linear boundary-value problems and to determine the hydrodynamic coefficients.Translated from Prikladnaya Mekhanika, Vol. 40, No. 10, pp. 37–77, October 2004.The study was partially sponsored by the German Research Fund (der Deutsche Forschungsgemeinschaft), Grant 436 UKR113/33/0-3.     

Double event-triggered leader-following consensus and fault detection for Lipschitz nonlinear multi-agent systems via periodic sampling strategy

This paper considers the problem of leader-following consensus and fault detection for a class of multi-agent systems with Lipschitz nonlinear dynamics. To reduce the amount of redundant information and avoid checking triggering conditions continually, this paper proposes an efficient network framework with a double periodic event-triggered mechanism. Based on the proposed framework, an improved fault detection observer and a consensus controller are designed. Then, the original problem is converted into a set of stability problems with constraints. According to Lyapunov–Krasovskii theorem and the free-weighting matrix technique, sufficient conditions for solving these stability problems are derived in the form of bilinear matrix inequalities (BMIs). Further, to eliminate the nonlinear terms of BMI and obtain optimal performance, two iterative algorithms based on linear matrix inequalities (LMIs) are developed. Two simulation examples are provided to verify the practicality and validity of the theoretical results.

     

Dynamics of a spacecraft with large flexible appendage constrained by multi-strut passive damper

This paper is concerned with the dynamics of a spacecraft with multi-strut passive damper for large flexible appendage.The damper platform is connected to the spacecraft by a spheric hinge,multiple damping struts and a rigid strut.The damping struts provide damping forces while the rigid strut produces a motion constraint of the multibody system.The exact nonlinear dynamical equations in reducedorder form are firstly derived by using Kane’s equation in matrix form.Based on the assumptions of small velocity and small displacement,the nonlinear equations are reduced to a set of linear second-order differential equations in terms of independent generalized displacements with constant stiffness matrix and damping matrix related to the damping strut parameters.Numerical simulation results demonstrate the damping effectiveness of the damper for both the motion of the spacecraft and the vibration of the flexible appendage,and verify the accuracy of the linear equations against the exact nonlinear ones.     

Dynamic analysis of nonlinear systems by modal synthesis techniques

Different kinds of modal synthesis method have been used widely in dynamic analysis of linear structure systems, but, in general, they are not suitable for nonlinear systems.In this paper, a kind of modal synthesis techniques is extended to dynamic analysis of nonlinear systems. The procedure is based upon the method suggested in [20],[21], which is applicable to vibration analysis for complex structure systems with coupling attachments but with simplified forms of linear springs and dampers. In fact, these attachments have nonlinear characteristics as those generally known to the cases of nonlinear elasticity and nonlinear damping, e.g., piecewise-linear springs, softening or hardening springs. Coulomb damping,elas-ioplastic hysteresis damping, etc. So long as the components of structure are still linear systems, we can get a set of independent free-interface normal mode information hut only keep the lower-order for each component. This can be done by computations or experiments or both. The global equations of linear vibration are set up by assembling of the component equations of motion with nonlinear coupling forces of attachments. Then the problem is reduced to less degrees of freedom for solving nonlinear equations. Thus considerable saving in computer storage and execution time can be expected. In the case of a very high-order system, if sufficient degrees of freedom are reduced, then it may be possible for the problem to be solved by the aid of a computer of ordinary grade.As the general nonlinear vibration of multiple degrees of freedom systems is quite involved, in general, the exact solution of a nonlinear system equations is not easy to find, so the numerical method can be adopted for solving the reduced nonlinear equations to obtain the transient response of system for arbitrary excitations.     

Control Problems for Systems with Uncertainty

A review is made of the results obtained at the S. P. Timoshenko Institute of Mechanics in synthesis of robust controllers for linear stationary and periodic systems with uncertainty. The emphasis is on algorithms for construction of the Lyapunov function by solving the matrix Riccati equation and linear matrix inequalities. To illustrate the algorithms proposed, stabilization systems with uncertainty are synthesized for an A4D aircraft and a hopping vehicle. The numerical implementation of the algorithms involves no difficulties, since they are oriented toward standard MATLAB routines     

Large deformation solution of stiffened plates by a mixed finite element method

In the present paper, a finite element mixed variational functional and the iterative equations of the eccentric orthogonal stiffened plates are developed in accordance with nonlinear elasticity. By using an important technique the coupling coefficients of the two-dimensional coupling matrix are resolved into the known input data in the programming which is a three-dimensional coefficient matrix. The nonlinear equations are transformed into the instantaneous linear equations; and by using the conjugate gradient method the linear equations are solved. As a result, therefore, the calculation is enormously simplified, the precision manifested, and a satisfactory result obtained.     

基于黎曼几何的非线性振子动力学分析递推解析算法

基于黎曼几何和变分原理,推导了黎曼流形上非线性耗散动力系统的二阶微分动力学方程,并运用流形收缩的概念将动力学方程离散化,进而建立了相应的递推求解格式。选取3个自治非线性阻尼振子系统,分别采用递推解析算法和龙格库塔法求解微分动力学方程,并比较分析了不同的时间步长下两种算法的计算耗时。结果表明,与龙格库塔法相比,基于黎曼几何的递推算法不仅能得到每一时步的解析表达式,而且计算耗时短,计算效率高。基于黎曼流形的动力学方程递推算法为非线性动力学系统的解析求解提供了新思路。     

精细积分方法的发展与扩展应用

钟万勰院士于1991年首先提出计算矩阵指数的精细积分方法,其要点是2^N类算法和增量存储。精细积分方法可给出矩阵指数在计算机意义上的精确解,为常微分方程的数值计算提供了高精度、高稳定性的算法,现已成功应用于结构动力响应、随机振动、热传导以及最优控制等众多领域。本文首先介绍矩阵指数精细积分方法的提出、基本思想和发展;然后依次介绍在时不变/时变线性微分方程、非线性微分方程以及大规模问题求解中发展起来的各种精细积分方法,分析了其优缺点和适用范围;最后介绍了精细积分方法的基本思想在两点边值问题、椭圆函数和病态代数方程等问题的扩展应用,进一步展示了该思想的特色。