与浓度相关的扩散系数计算——逼近法

本文探讨了用逼近法计算扩散系数和浓度的函数关系的方法. 逼近法克服了Boltzman-Matano 法的主要缺点. 采用逼近,可以按我们要求的精度求出扩散系数与浓度之间的准确关系.本文研究了扩散方程及其解的一一对应性, 证明了逼近法的可靠性. 用逼近法研究了玻璃中一价阳离子之间的相互扩散行为, 实验证明结果较好。 关键词：     

高斯分布扩散光波导折射率分布的多项式近似

本文用适当的多项式函数近似高斯折射率分布函数,利用Ikuno的结果,导出了扩散平面光波导导模有效折射率的更精确近似解公式.这个公式不仅简单便于计算,而且由它求得的导模有效折射率更接近光线方法的数值结果,其精度不仅优于无微扰的适当双曲正割分布近似,也优于微扰的适当双曲正割分布近似.同时也改善了较高阶导模的数据精度.     

与浓度相关的扩散系数计算——逼近法

本文探讨了用逼近法计算扩散系数和浓度的函数关系的方法。逼近法克服了Boltzmann-Matano法的主要缺点。采用逼近法,可以按我们要求的精度求出扩散系数与浓度之间的准确关系。本文研究了扩散方程及其解的一一对应性,证明了逼近法的可靠性。用逼近法研究了玻璃中一价阳离子之间的相互扩散行为,实验证明结果较好。 关键词：     

二阶常微分方程组边值问题的一种数值解法

本文用压缩变挟方法,求解了函数解随自变量增加而迅速增长的一类常微分方程的边值问题。结果表明,这种方法,可加快收敛速度。在奇点W=0附近,采用局部解析近似处理,得到了比较满意的数值结果。     

Fermi折射率分布光波导的模方程

对Fermi折射率分布的特性进行了分析,用光线理论推导出Fermi折射率平面光波导的近似超越模方程.理论分析和实验结果比较表明,Fermi折射率分布函数能较好地描述准阶跃渐变折射率.得到的近似模方程相当精确地决定了这类波导的模式特性,比阶跃折射率近似精确.     

弱导多层平面光波导辐射模的自由空间模近似

本文用自由空间模简化及近似表示弱导多层介质光波导的辐射模.只要辐射方向与介质界面的角度不是很小,近似程度很好.将其应用于求解光栅耦合器的辐射损耗问题,使计算公式简化,用自由空间模计算的结果与用辐射模计算的结果比较接近,也与其它近似方法进行了比较.     

弱导多层平面光波导辐射模的自由空间模近似

本文用自由空间模简化及近似表示弱导多层介质光波导的辐射模.只要辐射方向与介质界面的角度不是很小,近似程度很好.将其应用于求解光栅耦合器的辐射损耗问题,使计算公式简化,用自由空间模计算的结果与用辐射模计算的结果比较接近,也与其它近似方法进行了比较.     

基于局部优化方法的半导体激光器反源技术

为了解半导体激光器的传播模近场分布,利用远场分布的光强和逼近信赖域的局部优化方法,提出了一种新的计算传播模近场分布的方法.模拟实验表明,只需很少几个展开函数的线性组合,就能逼近源场函数,是一种稳定性好,精度较高的方法.     

相对论性电子束的类Tokamak平衡

本文对具有双δ函数速度分布的相对论性电子束,考察了其环形平衡性质。通过对反环径比展开,得到了关于磁通函数及电势的一级近似解析解。还讨论了电子漂移轨道的近似描述。 关键词：     

梯度单模光纤之间的能量耦合

本文用高斯-变型贝塞耳(Gaussian-modified Bessel)函数近似基模场计算两梯度单模光纤之间的能量耦台.给出了耦台系数公式和不同v值及指数折射率分布参数g的耦合系数随距离D的变化曲线.g=∞时,本文的结果和文献[5]结果与Snyder阶跃光纤的结果进行了比较,说明本文方法的近似程度较好.     

A simultaneous electronic transition of the oxygen-naphthalene complex

Thorne's extension of the Enskog theory to thermal conductivity and viscosity of binary hard sphere fluid mixtures is examined. It is shown that for molecules of the same size and different mass the expression for the thermal conductivity of such a mixture obtained by Longuet-Higgins, Pople and Valleau is identical to the collisional term in Thorne's equation arising from the locally maxwellian velocity distribution. Thorne's equations are evaluated by using the Percus-Yevick approximation for the contact radial distribution functions obtained by Lebowitz. The variations of the transport coefficients are examined as functions of pressure, composition and ratios of diameters and masses of the two species. Some comparison is made with experimental results.     

Thermocapillary deformation of a thin locally heated horizontal liquid layer

In this paper, steady thermocapillary flow in a thin horizontal layer of a viscous incompressible liquid with a free surface is considered. An axially symmetric steady problem with a localized thermal action on a horizontal liquid layer with a deformable free surface is solved in a thin-layer approximation. In addition to the thermocapillary effect, the model takes into account the capillary pressure caused by the free surface variable curvature and the convective mechanism of heat transfer in the liquid. Analytical expressions for the velocity vector components as functions of the liquid layer thickness and surface temperature are obtained. The free surface and velocity profiles caused by various kinds of heating are calculated. The influence of convective heat transfer on the flow pattern is analyzed.     

A new pressure-dominant approximation model for acoustic–structure interaction

A high accuracy approximation modeling approach for the acoustic–structure interaction problem with a shell structure is presented in this paper. The new approximation model aims to accurately reveal the relationship between pressure and velocity in the acoustic field. The main idea of this model is to separate the velocity terms into a combination of velocity and pressure by using a weighting parameter. The modal analysis was performed to find an appropriate weighting parameter for the new model for the spherical case. The stability range of the model is limited during this process. An approximation model was coupled with the equation of motion of a spherical shell to check the performance of the model. Responses of a spherical shell excited by a plane step wave and cosine-type incident pressure from the new model were compared to the exact solution and solutions from former approximation models such as Doubly Asymptotic Approximations (DAAs). The new proposed model can approximate high accuracy responses in both early and late time.     

Oscillator strengths in certain air molecule systems—I. The O2 Schumann-Runge system

The extended-basis molecular wave functions of Cade are utilized for the computation of the electronic transition moment in the AEVO approximation for the Schumann-Runge system. The calculation is carried out for both the dipole length and dipole velocity forms of the operator. The matrix elements are, to a good approximation, linear functions of internuclear separation in both forms. The electronic oscillator strength thus computed differs from the most recent experimental value by a factor of three. The introduction of the dipole centroid yields results differing by about 50 per cent from observed.     

An improved approximation of Bergmann's form for the Rayleigh wave velocity

In the present paper an improved approximation for the Rayleigh wave velocity in isotropic elastic solids is obtained using the method of least squares. It is of Bergmann’s form, i.e. the form of the ratio of two binomials. It is shown that this approximation is the best one of the Rayleigh wave velocity, in the sense of least squares, with respect to the class of functions whose elements are the ratio of two binomials. This approximation is much more accurate than Bergmann’s one. Its maximum percentage error is 10 times smaller than that of Bergmann’s. It is 7.6 times better than the one obtained recently by Royer and Clorennec [D. Royer, D. Clorennec, An improved approximation for the Rayleigh wave equation, Ultrasonics 46 (2007) 23–24]. An approximation of Bergmann’s form for the squared Rayleigh wave velocity is also derived and its maximum percentage error is 5 times smaller than that of Royer and Clorennec’s approximation. Some polynomial approximations with very high accuracy are also obtained.     

Wave-number-based assessment of the doubly asymptotic approximation. I. Frequency domain wet surface impedance

The doubly asymptotic approximation (DAA) is a canonical relationship for the interaction between surface normal velocity and pressure. Its validity for a slender hemicapped cylinder is examined by formulating a frequency domain version of DAA using the global basis functions employed in the wave-number-based formulation of the surface variational principle [K. Wu and J. H. Ginsberg, ASME J. Vib. Acoust. 120, 392-400 (1998)]. The wet surface impedance matrix, which relates the spectral representation of normal velocity to a corresponding representation of pressure, is obtained according to a second-order version of DAA and according to the surface variational principle. Comparison and interpretation of the results reveals that DAA fails to account for highlights associated with transition from supersonic to subsonic surface waves as the surface wavelength decreases with frequency held constant.     

A model of Jupiter’s magnetodisk

An axially symmetric equilibrium model of Jupiter’s magnetodisk is developed in the MHD approximation that takes the plasma corotation and the centrifugal force into account. The model is constructed for two cases: (1) the magnetodisk plasma is assumed to have a uniform temperature; (2) the plasma pressure is assumed to be an adiabatic function of density. Analytical expressions for the magnetic field, current density, and magnetodisk temperature and thickness distributions are obtained as functions of the system parameters, viz., the radial distribution of plasma pressure in the equatorial plane, the transverse magnetic field in the center of the layer, and the angular velocity of the plasma rotation.     

Structural, electronic and optical properties of high pressure stable phases of ZnTe

We have performed full potential linear augmented plane wave calculations to investigate the pressure induced phase transition in ZnTe. Total energies of three phases (zinc-blende, cinnabar and Cmcm) are calculated using density functional theory formalism under generalized gradient approximation and Engel-Vosko generalized gradient approximation for the exchange correlation potential approximation. The pressure stability corresponding to zinc-blende, cinnabar and Cmcm phases of ZnTe are computed. We find that cinnabar phase could be formed as a metastable phase by releasing pressure from the high pressure Cmcm phase. The obtained structural, electronic and optical results are compared with previous calculations and available experimental data. Overall good agreement is found.     

The effective interaction in a nuclear model

The effective interactionU belonging to a given subspace ofn unperturbed model functions can be related to the free nucleon-nucleon interaction by means of an exact integral equation. The conditions for a Born-v. Neumann expansion ofU and for an approximation by a slowly energy dependent interaction with adjustable parameters are investigated. It appears that effective interactions which have been used successfully in shell model calculations of bound state properties do not necessarily describe the actual dynamics of a nucleus.     

A pressure-correction scheme for convection-dominated incompressible flows with discontinuous velocity and continuous pressure

In this work we present a pressure-correction scheme for the incompressible Navier–Stokes equations combining a discontinuous Galerkin approximation for the velocity and a standard continuous Galerkin approximation for the pressure. The main interest of pressure-correction algorithms is the reduced computational cost compared to monolithic strategies. In this work we show how a proper discretization of the decoupled momentum equation can render this method suitable to simulate high Reynolds regimes. The proposed spatial velocity–pressure approximation is LBB stable for equal polynomial orders and it allows adaptive p-refinement for velocity and global p-refinement for pressure. The method is validated against a large set of classical two- and three-dimensional test cases covering a wide range of Reynolds numbers, in which it proves effective both in terms of accuracy and computational cost.