Green's functions for a piezoelectric semi-infinite body with a fixed conductor surface

By using Stroh's formalism, simple explicit compact expressions of Green's functions for a piezoelectric semi-infinite body, with a fixed conductor surface electrode, subject to a singularity (i.e., a generalized line dislocation and a generalized line force at a point z°) are presented. Coulomb forces acted on the free line charge at z° due to the boundary polarization charges of the medium and the induction charges of the conductor together with the electromechanical coupling effects inside the region are analyzed in detail. The obtained results are valid not only for plane and antiplane problems but also for the coupled problems between inplane and outplane deformations.     

A hybrid/mixed model finite element analysis for eigenvalue problem of moderately thick plates

The buckling and free vibration problems of moderately thick plate are considered in this paper by using the hybrid/mixed finite element model. A modified Reissner principle which only requires C⁰ continuity is derived. No lockling phenomenon is observed. Linear interpolation is used for all independent unknown function. Finally a displacement generalized eigenvalue equation is obtained, in which the stiffness matrix is symmetric and positively definite. The calculated results show that the method proposed is simple, reliable and satisfactory.     

改进型XFEM进展

扩展有限元法(XFEM)在诞生后的十几年时间里,引起学术界和工业界的广泛关注,并已经成为目前裂纹分析的主流数值方法。然而,在实际应用中该法一直受到两方面的困扰,(1)总体方程高度病态;刚度阵条件数随网格尺寸呈h-6变化(普通有限元为h-2)。(2)裂尖强化插值由于能量一致性问题无法直接推广应用于动力学计算。前者表现在XFEM稳态问题的迭代求解收敛慢或难以收敛,后者长期以来导致XFEM裂纹扩展动力学计算实施困难。本文认为XFEM目前遇到的种种困难,均与单位分解引入的额外自由度相关。为此,提出了无额外自由度的单位分解插值格式,基于此格式,进一步构造出改进型扩展有限元方法。改进型XFEM具有如下特点,(1)可以消除原有XFEM的线性依赖性和总体方程病态的问题。(2)避免动力学问题中额外自由度引起的质量集中、零临界时间步长问题以及裂纹扩展过程中的能量一致性问题。本文结合静动力学测试问题综述上述改进。     

Applying the generalized step-function to solve the bending problems of nonuniform beams with variable cross section

The bending problems of nonuniform beams with variable cross-section can be approximated by that of a step beam under sectionally uniform load (including both concentrated forces and couples). In this paper, the concept of Heaviside function {x-a}⁰ will be generalized, and a new function {x-a}⁰, n=0,1,2…,will be defined, which may be named as a generalized step function. The rules of operation will also be given to {x-a}ⁿ{x-b}⁰. The reciprocal of the flexural of rigidity 1/EJ and the bending moment M(x) can all be expressed in terms of {x-a}ⁿ,and substituted into the differential equation of the elastic curve of the beam respectively. Thus we may establish a set of unified method to solve various types of bending problems of straight beams. The general solution of the deflection equation will be given.     

An Experimental Investigation of Two-Point Correlations in Two- and Three-Dimensional Turbulent Boundary Layers

Two-point correlation measurements of the wall normal fluctuating velocities were made in two-dimensional (2-D) and pressure-driven three-dimensional (3-D) turbulent boundary layers. These data are needed for characterization and modeling of active-motion length scales, especially for 3-D flows. The fine-probe-volume data were measured using two custom-designed laser-Doppler-velocimeter fiber-optic probes. The data are relatively free of noise, signal broadening, and bias effects. Favorable comparisons with direct-numerical-simulation (DNS) results in the near-wall region of the 2-D flow validate the experimental techniques used here. For a given fixed probe location, non-dimensional correlation values scale best on the probe separation. For both the 2-D and 3-D cases, peak correlations lie along a line inclined away from the wall at 11° and 8°, respectively, which suggests the existence of an outgoing characteristic line affected by only the upstream flow. The decay of the correlation coefficient occurs nearer the wall than away from the wall relative to the fixed probe location. The variations for the 3-D flow correlations are similar to the 2-D variations, but with longer Δ x ⁺and Δ y ⁺ decay distances, probably because of the 3-D flowacceleration. While the spanwise variation of the correlationcoefficients is symmetric about the fixed point for the 2-D case asdictated by reciprocity, the 3-D case shows a large asymmetry for spanwise variations Δ z ⁺ < 68. The profiles at higher Δ z ⁺ are more symmetric. In general, at a given y the maximum correlation is skewed toa non-zero Δ z. It appears that the skewing of the correlation coefficient in the z direction tracks the sign of . This revised version was published online in July 2006 with corrections to the Cover Date.     

Systems of generalized vector quasi-variational inclusions and systems of generalized vector quasi-optimization problems in locally <Emphasis Type="Italic">FC</Emphasis>-uniform spaces

In this paper, we introduce some new systems of generalized vector quasi-variational inclusion problems and system of generalized vector ideal （resp., proper, Pareto, weak） quasi-optimization problems in locally FC-uniform spaces without convexity structure. By using the KKM type theorem and Himmelberg type fixed point theorem proposed by the author, some new existence theorems of solutions for the systems of generalized vector quasi-variational inclusion problems are proved. As to its applications, we obtain some existence results of solutions for systems of generalized vector quasi-optimization problems.     

Fixed pansystems theorems and pansystems logic conservation of chaos

The study of nonlinear problems was developed in works ^[1,2] by means of the pansystems methodology (PM) which does not need the condition related to differential manifolds and linear space. In work [2], within the framework of PM, we proved that study of panchaos, panattractor and strange panattractor can be transformed conditionally into some forms of fixed subsets. As a continuation of work [2], we now research the pansystems logic conservation of panchaos, panattractor, strange panattractor and some other fixed subsets.     

条带域内二维各向异性压电介质的Green函数

以压电各向异性弹性介质广义平面变形的Stroh一般解为基础，采用复变函数方法（即保角变换技术），研究了条带域介质内物理场的封闭形式解，求得了介质内某一点同时存在广义线位错和广义线力作用时的简单明确解，它就是边界元法中的Green函数，还分析了极化介质表面的电荷分布情况，并进而讨论了线电荷q与边界分布电荷间的库仑力问题，文中结果不仅适用于平面或反平面变形问题，而且也适用于两者耦合的二维变形问题。     

Removing Collision Singularities from Action Minimizers for the N-Body Problem with Free Boundaries

For the planar and spatial N-body problems, it has been proved by Marchal and Chenciner that solutions for the minimizing problem with fixed ends are free from interior collisions. This important result has been extended by Ferrario & Terracini to Newtonian-type problems and equivariant problems. It has also been used to construct many symmetric solutions for the N-body problem. In this paper we are interested in action minimizing solutions in function spaces with free boundaries. The function spaces are imposed with boundary conditions, such that every mass point starts and ends on two transversal proper subspaces of ℝ^d, d≥2. We will prove that solutions for this minimizing problem with free boundaries are always free from collisions, including boundary collisions. This result can be used to construct certain classes of relative periodic solutions of the N-body problem.     

Some problems in the Z-C-X space

The new concepts of the Z-C-X space and excellent cone are introduced. Some problems of random semiclosed 1-set-contractive operator are investigated in the Z-C-X space. At first, an important inequality is proved. Secondly, several new conclusions are proved by means of random fixed point index in the theory of random topological degree. A random solution of a class of random operator equations under conditions of imitating the parallelogram law is obtained, famous Altman’s theorem is generalized in partially ordered Z-C-X space. Therefore, some new results are obtained.     

Degree theory for multivalued (S) type mappings and fixed point theorems

The main purpose of this paper is devoted to generalizing the results of Browder ^[1,2]. This paper consists of four parts. In the first part, we introduce the concepts of multivalued (S) and (S) type mappings and the concepts of the limits of multivalued (S) and (S) type mappings. These kinds of mappings contain many monotone type mappings, such as maximal monotone mapping, bounded pseudo-monotone mapping and bounded generalized pseudo-monotone mapping, as its special cases. In the second part we define the pseudo-degree for (S) type mapping and the degree for (S) ₊ type mapping. These two kinds of degrees are all the generalizations of the degree defined by Browder ^[1,2].As applications, we utilize the degree theory presented in part 2 to study the existence of solutions for the multivalued operator equations (see part 3) and to obtain some new fixed point theorems in part 4.Research Supported by the National Natural Science Founation of China.     

Numerical analysis and accurate computation of heteroclinic orbits in the case of center manifolds

In earlier papers (see the preceding paper and the references there), Doedel and the author have developed a numerical method for the computation of branches of heteroclinic orbits for a system of autonomous ordinary differential equations in ⁿ in the case that the solution approaches the fixed points exponentially. The idea of the method is to reduce a boundary value problem on the real line to a boundary value problem on a finite interval by using linear approximation of the unstable and stable manifolds. Using the fact that the linearized operator of the problem is Fredholm in Banach spaces with exponential weights, the authors employed the general theory of approximation of nonlinear problems to show that the errors in the approximate solution decay exponentially with the length of the approximating interval. In this paper we extend the analysis in the preceding paper to the case of center manifolds which requires the refinement of the analysis in the preceding paper. The algorithm is applied to a model problem: the DC Josephson Junction. Computations are done using the software package AUTO.     

A method for establishing generalized variational principle

A method for establishing generalized variational principle is proposed in this paper. It is based on the analysis of mechanical meaning and it can be applied to problems in which the variational principles are needed but no corresponding variational principle is available. In this paper, the Hu-Washizu ’s generalized variational principle and the Hu’s generalized principle of complementary energy are derived from the mechanical meaning instead of from the generalization of the principle of minimum potenlial energy and the correct proofs of these two generaleed variational principles are given. It is also proved that this is wrong if one beleives that σ_ij, e_ij and ui are independent variables each other based on the reason that these three kinds of variables are all contained in these two generalized variational principles. The condition of using these two variational principles in a correct manner is also explained.     

Prediction of natural convection flow using network model and numerical simulations inside enclosure with distributed solid blocks

Steady state natural convection of a fluid with Pr ≈ 1 within a square enclosure containing uniformly distributed, conducting square solid blocks is investigated. The side walls are subjected to differential heating, while the top and bottom ones are kept adiabatic. The natural convection flow is predicted employing the nondimensional volumetric flow rate (Q_max^* Q_{\max }^{*} ) by using a network model and also using numerical simulations. For identical solid and fluid thermal conductivities (i.e. k _s  = k _f ), a parametric study of the effect of number of blocks (N ²), gap size (δ) and enclosure Rayleigh number (Ra) on Q_max^* Q_{\max }^{*} is performed using the two approaches. Network model predictions are observed to agree well with that from the simulations until Raδ³ ~ 12. Considering the enclosure with blocks as a porous medium, for a fixed enclosure Ra number, increasing the number of blocks for a fixed volumetric porosity leads to a decrease in enclosure permeability, which in turn reduces the flow rate. When the number of blocks is fixed, and for a given Ra number, the flow rate increases as the porosity increases by widening the gap between the blocks.     

Fourth‐order exponential finite difference methods for boundary value problems of convective diffusion type

Methods based on exponential finite difference approximations of h⁴ accuracy are developed to solve one and two‐dimensional convection–diffusion type differential equations with constant and variable convection coefficients. In the one‐dimensional case, the numerical scheme developed uses three points. For the two‐dimensional case, even though nine points are used, the successive line overrelaxation approach with alternating direction implicit procedure enables us to deal with tri‐diagonal systems. The methods are applied on a number of linear and non‐linear problems, mostly with large first derivative terms, in particular, fluid flow problems with boundary layers. Better accuracy is obtained in all the problems, compared with the available results in the literature. Application of an exponential scheme with a non‐uniform mesh is also illustrated. The h⁴ accuracy of the schemes is also computationally demonstrated. Copyright © 2001 John Wiley & Sons, Ltd.     

Extended Lagrangian formalism and the corresponding energy relations

In this paper an extended Lagrangian formalism for the rheonomic systems with the nonstationary constraints is formulated, with the aim to examine more completely the energy relations for such systems in any generalized coordinates, which in this case always refer to some moving frame of reference. Introducing new quantities, which change according to the law τ_a=φ_a(t), it is demonstrated that these quantities determine the position of this moving reference frame with respect to an immobile one. In the transition to the generalized coordinates qⁱ they are taken as the additional generalized coordinates q^a=τ_a, whose dependence on time is given a priori. In this way the position of the considered mechanical system relative to this immobile frame of reference is determined completely.Based on this and using the corresponding d'Alembert–Lagrange's principle, an extended system of the Lagrangian equations is obtained. It is demonstrated that they give the same equations of motion qⁱ=qⁱ(t) as in the usual Lagrangian formulation, but substantially different energy relations. Namely, in this formulation two different types of the energy change law dE/dt and the corresponding conservation laws are obtained, which are more general than in the usual formulation. So, under certain conditions the energy conservation law has the form E=T+U+P=const, where the last term, so-called rheonomic potential expresses the influence of the nonstationary constraints.Afterwards, a detailed analysis of the obtained results and their connection with the usual formulation of mechanics are given. It is demonstrated that so formulated energy relations are in full accordance with the corresponding ones in the usual vector formulation, when they are expressed in terms of the rheonomic potential. Finally, the obtained results are illustrated by several simple, but characteristic examples.     

Measurements of turbulence characteristics in an open-channel flow over a transitionally-rough bed using particle image velocimetry

The particle image velocimetry technique was used to measure characteristics of a turbulent flow over a transitionally-rough fixed bed in an open-channel flow. These conditions are typical of flows encountered in sediment transport problems. Measurements obtained with this technique were used to investigate the distributions of velocities, turbulence intensities, Reynolds stress, and third- and fourth-order moments in a region above y ⁺ = 10. The present results are in good agreement to those previously obtained on smooth walls and provide further evidence that PIV can be applied successfully to investigate turbulence in open-channel flows over a rough bed.     

p-version least squares finite element formulation for two-dimensional,incompressible fluid flow

A p-version least squares finite element formulation for non-linear problems is applied to the problem of steady, two-dimensional, incompressible fluid flow. The Navier-Stokes equations are cast as a set of first-order equations involving viscous stresses as auxiliary variables. Both the primary and auxiliary variables are interpolated using equal-order C⁰ continuity, p-version hierarchical approximation functions. The least squares functional (or error functional) is constructed using the system of coupled first-order non-linear partial differential equations without linearization, approximations or assumptions. The minimization of this least squares error functional results in finding a solution vector {δ} for which the partial derivative of the error functional (integrated sum of squares of the errors resulting from individual equations for the entire discretization) with respect to the nodal degrees of freedom {δ} becomes zero. This is accomplished by using Newton's method with a line search. Numerical examples are presented to demonstrate the convergence characteristics and accuracy of the method.     

An experimental study of fluctuating lift on a square-section cylinder oscillating transversely in a uniform stream

Experiments on a square-section cylinder fixed and forced to oscillate transversely in a uniform stream were conducted in a water tank. The Reynolds number of the experiments is in the range of 3·10³ to 10⁴, the amplitude to side length ratioA/D is up to 0.7 and the range of reduced velocity is 4.5<V _r <12. This study aims at investigating the lock-in phenomenon, the fluctuating lift and the phase shift between fluctuating lift and displacement of the oscillating cylinder. The problems on the aeroelastic instability relating to present experimental results have been discussed. The flow visualization clearly shows that there are drastic changes of vortex-shedding from cylinder at the resonance point and the upper end of the lock-in range. The results of the flow visualization give better understanding of the physical mechanism of the phase shift. Project supported by National Natural Science Foundation of China     

Orbits Homoclinic to Exponentially Small Periodic Orbits for a Class of Reversible Systems. Application to Water Waves

In this paper a class of reversible analytic vector fields is investigated near an equilibrium. For these vector fields, the part of the spectrum of the differential at the equilibrium which lies near the imaginary axis comes from the perturbation of a double eigenvalue 0 and two simple eigenvalues , . In the first part of this paper, we study the 4-dimensional problem. The existence of a family of solutions homoclinic to periodic orbits of size less than μ^N for any fixed N, where μ is the bifurcation parameter, is known for vector fields. Using the analyticity of the vector field, we prove here the existence of solutions homoclinic to a periodic orbit the size of which is exponentially small ( of order . This result receives its significance from the still unsolved question of whether there exist solutions that are homoclinic to the equilibrium or whether the amplitudes of the oscillations at infinity have a positive infimum. In the second part of this paper we prove that the exponential estimates still hold in infinite dimensions. This result cannot be simply obtained from the study of the 4-dimensional analysis by a center-manifold reduction since this result is based on analyticity of the vector field. One example of such a vector field in infinite dimensions occurs when describing the irrotational flow of an inviscid fluid layer under the influence of gravity and small surface tension (Bond number ) for a Froude number F close to 1. In this context a homoclinic solution to a periodic orbit is called a generalized solitary wave. Our work shows that there exist generalized solitary waves with exponentially small oscillations at infinity. More precisely, we prove that for each F close enough to 1, there exist two reversible solutions homoclinic to a periodic orbit, the size of which is less than , l being any number between 0 and π and satisfying . (Accepted October 2, 1995)