Allowing for Rotations about the Normal in Nonlinear Theories of Shells

Consideration is given to three versions of nonlinear strain–displacement relations in the case of small strains and moderately small angles of rotation: (i) relations that neglect rotations about the normal in conformity with the hypotheses of the Donnel–Mushtary–Vlasov theory; (ii) relations, derived from the elasticity equations using Novozhilov's tensor, that exactly allow for rotations; and (iii) relations, proposed by Sanders, that allow for rotations but neglect shear strains. These versions are compared by comparing the solutions of the stability problem for a corrugated cylindrical shell. It is established that the critical loads are close when rotations are allowed for exactly and when Sanders' technique is used     

On the effective behavior of nonlinear inelastic composites: II: A second-order procedure

A new method for determining the overall behavior of composite materials comprising nonlinear viscoelastic and elasto-viscoplastic constituents is presented. Part I of this work showed that upon use of an implicit time-discretization scheme, the evolution equations describing the constitutive behavior of the phases can be reduced to the minimization of an incremental energy function. This minimization problem is rigorously equivalent to a nonlinear thermoelastic problem with a transformation strain which is a nonuniform field (not even uniform within the phases). In part I of this paper the nonlinearity was handled using a variational (or secant) technique. In this second part of the study, a proper modification of the second-order procedure of Ponte Castañeda is proposed and leads to replacing, at each time-step, the actual nonlinear viscoelastic composite by a linear viscoelastic one. The linearized problem is even further simplified by using an “effective internal variable” in each individual phase. The resulting predictions are in good agreement with exact results and improve on the predictions of the secant model proposed in part I of this paper.     

求非线性动力系统周期解的切比雪夫多项式法

周期运动是一种在客观世界中普遍存在的运动形式,它与混沌运动之间存在十分密切的关系,因而具有很重要的研究价值。利用切比雪夫多项式的若干良好性质,对自治非线性动力系统进行分析,将状态矢量在主周期上展开为切比雪夫多项式的形式,从而将原问题转变为非线性代数方程组的求解问题,得出一种可以方便、迅速地获得周期轨道近似多项式表达式的方法。此方法不依赖于小参数假设,可以用于分析强非线性问题,而且对参数激励系统同样有效。在计算机条件允许时,对高维系统也能迅速、精确地得到其周期轨道的近似多项式表达式。以三维Rossler系统和五维非线性磁浮转子系统周期轨道的计算为例,通过与四阶Runge-Kutta数值积分结果比较,说明此方法的精确、高效性。     

Geometrically exact curved beam element using internal force field defined in deformed configuration

This paper presents a study on the development of high-performance finite elements for geometrically nonlinear analysis of frame structures with curved members. Based on the geometrically exact beam theory, a highly efficient and accurate mixed finite element is developed. A new approach is proposed for constructing the independent internal force field by including major terms satisfying equilibrium conditions in the deformed configuration. An element-level equilibrium iteration procedure is employed for the condensation of element internal degrees of freedom during the nonlinear solution. Numerical results are presented to demonstrate the excellent performance of the element developed, and it is shown that even when each structural member is modelled with just one element, accurate solutions can still be achieved.     

A THEORETICAL MODEL FOR NONLINEAR ORBITAL MOTIONS OF ROTORS UNDER FLUID CONFINEMENT

In previous papers, Antunes and co-workers developed a theoretical model for nonlinear planar motions—motions X (t) taking place in one single direction—of rotors under fluid confinement using simplified flow equations on the gap-averaged fluctuating quantities. The nonlinear solution obtained was shown to be consistent with a linearized solution for the same problem. Also, it displayed an encouraging qualitative agreement between the nonlinear theory and preliminary experimental results. Following a similar approach, the nonlinear theoretical model is here extended to cope with orbital rotor motions—motions X (t) and Y (t) taking place in two different orthogonal directions—by developing an exact formulation for the two- dimensional dynamic flow forces. Numerical simulations of the nonlinear rotor–flow coupled system are presented and compared with the linearized model. These yield similar results when the eccentricity and the spinning velocity are low. However, if such conditions are not met, the qualitative dynamics stemming from the linearized and nonlinear models may be quite distinct. Preliminary experimental results also indicate that the nonlinear flow model leads to better predictions of the rotor dynamics when the eccentricity is significant, when approaching instability, and for linearly unstable regimes.     

Postbuckling Analysis of Nonconservative Elastic Systems

Abstract

Previous work on the postbuckling and imperfection-sensitivity of elastic structures has concentrated on conservative systems. The results of Koiterand others have led to a general theory of nonlinear stability behavior for these systems. The theory must be modified when nonconservative forces are present, and this is the aim of the present paper.

Discrete, nonconservative, elastic systems which exhibit static (divergence) instability are considered. The nonlinear behavior in the neighborhood of a critical point is analyzed by means of a perturbation procedure. When the critical point is simple, the results are similar to those for conservative systems. When a coincident critical point exists, however, different types of behavior occur. In many cases there is no bifurcation at all, with only the fundamental (trivial) equilibrium path passing through the critical point. Imperfection-sensitivity is more severe than for the typical bifurcation points and can even occur when the perfect system has no bifurcation. The results are illustrated with the use of a nonlinear double pendulum model subjected to a partial follower load.     

Nonlinear airship aeroelasticity

The aeroelastic derivatives for today's aircraft are calculated in the concept phase using a standard procedure. This scheme has to be extended for large airships, due to various nonlinearities in structural and aerodynamic behaviour. In general, the structural model of an airship is physically as well as geometrically nonlinear. The main sources of nonlinearity are large deformations and the nonlinear material behaviour of membranes. The aerodynamic solution is also included in the nonlinear problem, because the deformed airship influences the surrounding flow. Due to these nonlinearities, the aeroelastic problem for airships can only be solved by an iterative procedure. As one possibility, the coupled aerodynamic and structural dynamic problem was handled using linked standard solvers. On the structural side, the Finite-Element program package ABAQUS was extended with an interface to the aerodynamic solver VSAERO. VSAERO is based on the aerodynamic panel method using potential flow theory. The equilibrium of the internal structural and the external aerodynamic forces leads to the structural response and a trimmed flight state for the specified flight conditions (e.g. speed, altitude). The application of small perturbations around a trimmed state produces reaction forces and moments. These constraint forces are then transferred into translational and rotational acceleration fields by performing an inertia relief analysis of the disturbed structural model. The change between the trimmed flight state and the disturbed one yields the respective aeroelastic derivatives. By including the calculated derivatives in the linearised equation of motion system, it is possible to judge the stability and controllability of the investigated airship.     

Three-dimensional surface waves propagating over long internal waves

A non-uniform current, such as may be generated by long internal waves, interacts with short surface waves and causes patterns on the sea surface that are of interest. In particular, regions of steep breaking waves may be relevant to specular radar scattering.A simple approach to modelling this problem is to take a set of short, surface waves of uniform wavenumber on the sea surface, as may be caused by a gust of wind. The direction of propagation of the surface waves is firstly taken to be the same as that of the current, and surface tension and viscous effects are neglected. We have a number of methods of solution at our disposal: linear (one-dimensional) ray theory is simple to apply to the problem, a nonlinear Schrödinger equation for the modulated wave amplitude, modified to include to effect of the current, can be used and solutions can be found using a fully nonlinear irrotational flow solver. Comparisons between the ‘exact’ nonlinear calculations for two dimensions (which are too complicated/ computationally intensive to be extended to three dimensions) compare well with the two approximate methods of solution, both of which can be extended, within their limitations, to model the full three-dimensional problem; here we present three-dimensional results from the linear ray theory.By choosing such a simple (although we consider physically realistic) initial state of uniform wavenumber short waves and assuming a sinusoidal surface current, we can reduce the two-dimensional problem to dependence on three non-dimensional parameters.In three-dimensions, we consider an initial condition with a uniform wavetrain at an angle α say, to the propagating current, thus introducing a fourth parameter into the problem. Extension of the linear ray theory from one space to two space dimensions is numerically quite simple since we maintain uniformity in the direction perpendicular to the current, and the only difficulty lies with the presentation of results, due to the large number of variables now present in the problem such as initial wavenumber, angle of propagation, position in (x, y, t) space etc. In this paper we present just one solution in detail where waves are strongly refracted and form two distinct foci in space-time. There is a collimation of the short waves with the direction of the propagating current.     

A justification of nonlinear properly invariant plate theories

A single asymptotic derivation of three classical nonlinear plate theories is presented in a setting which preserves the frame-invariance properties of three-dimensional finite elasticity. By a successive scaling of the external loading on the three-dimensional body, the nonlinear membrane theory, the nonlinear inextensional theory and the von Kármán equations are derived as the leading-order terms in the asymptotic expansion of finite elasticity. The governing equations of the nonlinear inextensional theory are of particular interest where 1) plane-strain kinematics and plane-stress constitutive equations are derived simultaneously from the asymptotic analysis, 2) the theory can be phrased as a minimization problem over the space of isometric deformations of a surface, and 3) the local equilibrium equations are identical to those arising in the one-director Cosserat shell model. Furthermore, it can be concluded that with a regular, single-scale asymptotic expansion it is not possible to obtain a system of plate equations in which finite membrane strain and finite bending strain occur simultaneously in the leading-order term of an asymptotic analysis.     

Invariant Manifolds for Steady Boltzmann Flows and Applications

We develop a theory of invariant manifolds for the steady Boltzmann equation and apply it to the study of boundary layers and nonlinear waves. The steady Boltzmann equation is an infinite dimensional differential equation, so the standard center manifold theory for differential equations based on spectral information does not apply here. Instead, we employ a time-asymptotic approach using the pointwise information of Green’s function for the construction of the linear invariant manifolds. At the resonance cases when the Mach number at the far field is around one of the critical values of ?1, 0 or 1, the truly nonlinear theory arises. In such a case, there are wave patterns combining the fast decaying Knudsen-type and slow varying fluid-like waves. The key Knudsen manifolds consisting of only Knudsentype layers are constructed through delicate analysis of identifying the singular behavior around the critical Mach numbers. Around Mach number ± 1, the fluidlike waves are compressive and expansive waves; and around the Mach number 0, they are linear thermal layers. The quantitative analysis of the fluid-like waves is done using the reduction of dimensions to the center manifolds.Two-scale nonlinear dynamics based on those on the Knudsen and center manifolds are formulated for the study of the global dynamics of the combined wave patterns. There are striking bifurcations in the transition of evaporation to condensation and in the transition of the Milne’s problem with a subsonic far field to one with a supersonic far field. The analysis of these wave patterns allows us to understand the Sone Diagram for the study of the complete condensation boundary value problem. The monotonicity of the Boltzmann shock profiles, a problem that initially motivated the present study, is shown as a consequence of the quantitative analysis of the nonlinear fluid-like waves.     

Solution of the first boundary value problem of nonlinear theory of steady-state creep for a half-space in antiplane deformation

The first boundary value problem of nonlinear theory of steady-state creep with powerlaw relationship between the stresses and the strain rates is considered for a half-space under conditions of antiplane (out-of-plane) deformation when tangential distributed forces are given on the half-space boundary. By using the introduced harmonic pseudostress function, we reduce solving this problem to solving a nonlinear singular integral equation admitting an exact solution.     

非线性模态的分类和新的求解方法

引入不可分偶数维不变流形的概念来定义非线性模态．在此基础上，揭示出了一种新的模态——耦合非线性模态，并对实际系统中各种可能的模态进行了分类．这种分类可能是新的构筑非线性模态理论的框架．用此方法构造非线性模态，得到的模态振子具有范式的形式，形式最简、却能反映原系统在平衡点附近的主要动力学行为，且易于得到非线性频率及非线性稳定性等方面的信息．不仅适用于分析一般的多自由度系统，还可用于分析奇数维系统；不仅可构造内共振系统的非耦合模态，还可用于构造内共振耦合模态．从掌握的资料看，以前的方法还不能解决上述所有问题     

最优轨迹规划方法及其平动点航天器编队均衡耗能重构应用

由于均衡耗能航天器编队能够提高整体航天器编队服役时间,针对平动点航天器编队重构的均衡耗能最优轨迹规划问题,提出一种以状态、协态和控制三类变量插值为核心的求解非线性最优控制问题的新方法。基于连续时间表达的非线性最优控制问题通过变分原理转化为非线性方程组的求解,并进一步推导非线性方程组显示格式的Jacobi矩阵提高非线性方程组的计算效率。本文方法既满足最优控制理论的一阶必要条件又具有较大的收敛域;同时,不需要对协态初值准确猜测,避免了大规模非线性规划问题的求解。通过对中心航天器位置固定和无中心航天器两种情况的数值模拟,结果表明,本文方法对航天器编队重构轨迹规划问题能够达到均衡耗能的目标,具有一定的应用价值。     

带有轴间动力学解耦的三轴转台自适应控制

针对某型号转台存在的轴间非线性耦合，运用非线性解耦的方法进行了解耦设计，并基于解耦后的系统提出用超稳定性和正性的方法进行自适应模型跟随控制器设计，来解决系统在运行过程中参数发生变化对控制性能影响的问题。仿真结果表明，当系统参数在一定范围变化甚至被控对象模型阶次发生变化时，所设计的自适应模型跟随控制器仍能保证系统具有良好的控制性能。     

微尺度泊肃叶流的高阶连续模型分析

分别从分子运动论及连续流理论出发,对体积力驱动的微尺度平面泊肃叶(Poiseuille)流的横向分布特征进行了分析. 分子水平模拟采用直接模拟蒙特卡罗(direct simulation Monte Carlo, DSMC)方法;连续流理论则主要考察了伯内特(Burnett)及超伯内特(Super-Burnett)等高阶连续模型,在平行流假设下,获得一组高阶非线性常微分方程,补充完整的边界条件,并应用龙格-库塔(Runge-Kutta)方法求解. 结果表明,即使对于过渡领域流动,高阶连续模型可以给出与DSMC 结果完全相符的压力分布,而速度分布当努森(Knudsen)数约为0.2时即在壁面开始出现偏差;对于温度的横向分布,伯内特模型回复到纳维-斯托克斯(Navier-Stokes)水平,不能得到与DSMC一致的双峰结构,而超伯内特模型在滑移流动领域与DSMC定性相符,在过渡领域却仅能正确预测主流区温度分布,壁面附近差异明显;横向热流与纳维-斯托克斯模型预测接近,但机理上存在本质区别. 本文结果提示选用连续模型时,不仅要根据流动参数来判断,还可以根据所关注的物理量来进行调整,适度扩大连续模型的适用范围. 但即使采用高阶本构关系,连续模型仍然不能完全描述壁面附近区域的非平衡效应(如努森层效应),这是试图扩大连续模型适用范围时必然会遇到的困难.     

基于多种搜索策略的人工蜂群算法的结构损伤识别

人工蜂群算法是一种元启发式算法,具有构架简单,易于操作和鲁棒性较好的特点。本文对原始蜂群算法进行了改进,为蜜蜂们提供了更丰富的搜索策略。基于轮盘赌原则选择更优的迭代方式,进而使算法的收敛速度和精度有了显著的提高。基于频率残差和模态置信准则(MAC)建立结构损伤识别问题的目标函数,然后利用ABC算法,QABC算法和本文方法求解该目标函数,得到损伤识别的结果。利用一桁架结构做数值模拟,用简支梁结构进行实验验证。算例表明,改进的算法更能有效地检测出结构的局部损伤,具有对测量噪声不敏感、高效率以及高精度等优点。     

边界积分方程中近奇异积分计算的一种变量替换法

准确估计近奇异边界积分是边界元分析中一项很重要的课题,其重要性仅次于对奇异积分的处理. 近年来已发展了许多方法,都取得了一定程度的成功,但这个问题至今仍未得到彻 底的解决. 基于一种新的变量变换的思想和观点,提交了一种通用的积分变换法, 它非常有效地改善了被积函数的震荡特性,从而消除了积分的近奇异性,在不增加计算量的情况 下, 极大地改进了近奇异积分计算的精度. 数值算例表明,其算法稳定,效率高, 并可达到很高的计算精度,即使区域内点非常地靠近边界,仍可取得很理想的结果.     

破碎岩体非等温渗流的非线性动力学研究

分别从固体及流体导热的能量方程出发,导出破碎岩体非等温渗流的能量本构方程, 结合渗流的连续性方程、运动方程、状态方程等建立了破碎岩体非等温渗流的一维非线性动力学方程组;结合Mathcad软件计算得到了系统的无量纲化平衡态, 利用逐次亚松弛迭代法分析了对应于不同参数时平衡态的稳定性;指出非等温渗流系统存在鞍结分岔及折叠突变, 与等温渗流相比, 考虑温度场的破碎岩体渗流动力系统更容易发生渗流突变.      

基于特征正交分解的一类瞬态非线性热传导问题的新型快速分析方法

提出了一种基于特征正交分解(POD)和有限元法的瞬态非线性热传导问题的模型降阶快速分析方法, 建立了导热系数随温度变化的一类瞬态非线性热传导问题有限元格式的POD降阶模型. 在隐式时间推进方法的基础上有效结合单元预转换方法和多级线性化方法发展了一种加速求解瞬态非线性热传导降阶模型的新型计算方法,并通过二维和三维算例验证了该方法的准确性和高效性. 研究结果表明: (1)降阶模型解的均方根误差在经过初始时段轻微的脉动后稳定于0.01%以下, 而其计算效率比有限元全阶模型提高2$\sim $3个数量级, 并且自由度数量(DOFs)愈大提高的幅度也愈加显著; (2)新型算法解决了常规算法在计算非线性降阶模型时加速性能差的问题, 即使是在DOFs比较小的时候也能够明显提高计算效率; (3)常数边界条件下得到的POD模态可以用来建立相同求解域在各种复杂时变边界条件下的瞬态非线性热传导降阶模型, 并对其传热过程和温度场进行快速准确的分析与预测, 具有很好的工程应用价值.      

Geometric Optics and Instability for Semi-Classical Schrödinger Equations

We prove some instability phenomena for semi-classical (linear or) nonlinear Schrödinger equations. For some perturbations of the data, we show that for very small times, we can neglect the Laplacian, and the mechanism is the same as for the corresponding ordinary differential equation. Our approach allows smaller perturbations of the data, where the instability occurs for times such that the problem cannot be reduced to the study of an ordinary differential equation.