Fourier spectral embedded boundary solution of the Poisson’s and Laplace equations with Dirichlet boundary conditions

A Fourier spectral embedded boundary method, for solution of the Poisson’s equation with Dirichlet boundary conditions and arbitrary forcing functions (including zero forcing function), is presented in this paper. This iterative method begins by transformation of the Dirichlet boundary conditions from the physical boundaries to some corresponding regular grid points (which are called the numerical boundaries), using a second order interpolation method. Then the transformed boundary conditions and the forcing function are extended to a square, smoothly and periodically, via multiplying them by some suitable error functions. Instead of direct solution of the resulting extended Poisson’s problem, it is suggested to define and solve an equivalent transient diffusion problem on the regular domain, until achievement of the steady solution (which is considered as the solution of the original problem). Without need of any numerical time integration method, time advancement of the solution is obtained directly, from the exact solution of the transient problem in the Fourier space. Consequently, timestep sizes can be chosen without stability limitations, which it means higher rates of convergence in comparison with the classical relaxation methods. The method is presented in details for one- and two-dimensional problems, and a new emerged phenomenon (which is called the saturation state) is illustrated both in the physical and spectral spaces. The numerical experiments have been performed on the one- and two-dimensional irregular domains to show the accuracy of the method and its superiority (from the rate of convergence viewpoint) to the other classical relaxation methods. Capability of the method, in dealing with complex geometries, and in presence of discontinuity at the boundaries, has been shown via some numerical experiments on a four-leaf shape geometry.     

变系数瞬态热传导问题边界元格式的特征正交分解降阶方法

提出了一种基于边界元法求解变系数瞬态热传导问题的特征正交分解(POD)降阶方法,重组并推导出变系数瞬态热传导问题适合降阶的边界元离散积分方程,建立了变系数瞬态热传导问题边界元格式的POD降阶模型,并用常数边界条件下建立的瞬态热传导问题的POD降阶模态,对光滑时变边界条件瞬态热传导问题进行降阶分析.首先,对一个变系数瞬态热传导问题,建立其边界域积分方程,并将域积分转换成边界积分;其次,离散并重组积分方程,获得可用于降阶分析的矩阵形式的时间微分方程组;最后,用POD模态矩阵对该时间微分方程组进行降阶处理,建立降阶模型并对其求解.数值算例验证了本文方法的正确性和有效性.研究表明:1)常数边界条件下建立的低阶POD模态矩阵,能够用来准确预测复杂光滑时变边界条件下的温度场结果;2)低阶模型的建立,解决了边界元法中采用时间差分推进技术求解大型时间微分方程组时求解速度慢、算法稳定性差的问题.     

任意入射辐射条件下介质特性对瞬态辐射传递的影响

用基于Monte Carlo法的DRESOR法在平行平板系统内具有吸收、无发射介质中研究不同波形入射、壁面反射、介质散射率、光学厚度、各向异性散射等条件对瞬态辐射传递的影响.任意连续波形入射辐射是目前大多数数值方法很难处理的瞬态辐射问题,而DRESOR法通过在系统内计算一单位入射辐射能对介质的DRESOR数分布,就能计算任意连续波形入射辐射条件下高方向分辨率的瞬态辐射强度结果.DRESOR法和Monte Carlo法计算的结果进行了比较验证,两者吻合较好,证明了DRESOR法处理瞬态入射辐射问题的正确性和有效性.     

Research on the nonlinear pulse propagation by numerical analysis

Based on the computation approaches of numerical analysis in electromagnetic, the nonlinear pulse propagation in optical fiber is investigated in this paper. The transforming rule of the shape and spectrum of Gauss pulse are also researched in chirped and non-chirped scenarios from the nonlinear Schrödinger equation (NLSE). Analyzing the transient wave behavior in time domain and frequency domain respectively, the compression effect of chirp is also presented at different propagation distance. Then by the nonlinear Schrödinger equation, we discuss a specific solution in the quasi soliton without distortion, which has the stable propagation properly, and do the numerical simulation in soliton wave. Finally, by selecting proper fiber structure parameters and adjusting the width in soliton pulse, the soliton communication system can be optimized.     

Effect of nuclear motion on molecular high-order harmonics and on generation of attosecond pulses in intense laser pulses

We calculate harmonic spectra and shapes of attosecond-pulse trains using numerical solutions of Non-Born-Oppenheimer time-dependent Shr?dinger equation for 1D H2 molecules in an intense laser pulse. A very strong signature of nuclear motion is seen in the time profiles of high-order harmonics. In general the nuclear motion shortens the part of the attosecond-pulse train originating from the first electron contribution, but it may enhance the second electron contribution for longer pulses. The shape of time profiles of harmonics can thus be used for monitoring the nuclear motion.     

Three Boundary Meshless Methods for Heat Conduction Analysis in Nonlinear FGMs with Kirchhoff and Laplace Transformation

This paper presents three boundary meshless methods for solving problems of steady-state and transient heat conduction in nonlinear functionally graded materials (FGMs). The three methods are, respectively, the method of fundamental solution (MFS), the boundary knot method (BKM), and the collocation Trefftz method (CTM) in conjunction with Kirchhoff transformation and various variable transformations. In the analysis, Laplace transform technique is employed to handle the time variable in transient heat conduction problem and the Stehfest numerical Laplace inversion is applied to retrieve the corresponding time-dependent solutions. The proposed MFS, BKM and CTM are mathematically simple, easy-to-programming, meshless, highly accurate and integration-free. Three numerical examples of steady state and transient heat conduction in nonlinear FGMs are considered, and the results are compared with those from meshless local boundary integral equation method (LBIEM) and analytical solutions to demonstrate the efficiency of the present schemes.     

聚合物驱试井分析叠加原理适用性的探索

建立考虑聚合物剪切变稀特性的聚驱试井数学模型,基于矩形网格和井筒周围径向加密的复合网格。采用有限体积方法求得数值解,将数值解与叠加原理得到的注聚井停注压降进行对比。结果表明：叠加原理计算得到的注聚井关井井底压力值远低于数值解,说明叠加原理不能用于聚合物驱试井解释。在实测注聚井压降资料解释中,叠加原理得到的渗透率明显偏低,进一步证明了叠加原理不适合非线性聚驱试井模型,数值方法更加适用于聚驱试井解释。研究结论适用于类似的非线性试井问题。     

运动声源识别的动态波叠加方法研究

为更好地识别运动声源并解决声源识别中的虚假声源问题,基于运动声源短时信号的Doppler频移特性, 建立运动声场的短时波叠加关系, 利用波束形成方法对声源点进行预估, 基于预估建立起多运动声源的动态叠加方程, 进一步通过波叠加方程的求解进行声源的计算, 从而创建一种可以用于运动声源识别的动态波叠加方法. 该方法可以有效识别运动声源, 将波叠加方法扩展到了运动声源测量领域, 并在不增加传声器数量以及改变阵列形式的条件下有效抑制运动声源重建中的旁瓣效应, 解决运动声源识别中的虚假声源问题. 仿真及实际运动声源的测量试验结果证明了该方法的有效性.     

An elastodynamic solution of finite long orthotropic hollow cylinder under torsion impact

The paper presents an analytical method to solve the elastodynamic problem of a finite-length orthotropic hollow cylinder subjected to a torsion impact often occurring in engineering fields. The elastodynamic solution is composed of a quasi-static solution of homogeneous equation satisfied with the non-homogeneous boundary condition and a dynamic solution of non-homogeneous equation satisfied with homogeneous boundary condition. The quasi-static solution can be obtained by directly solving the quasi-static equation satisfied with the non-homogeneous boundary condition. The solution of a non-homogeneous dynamic equation is obtained by means of a finite Hankel transform to a radial variable r, Laplace transform to a time variable t and finite Fourier transform to an axial variable z. Thus, the elastodynamic solution of the finite length of an orthotropic hollow cylinder subjected to a torsion impact is obtained. On the other hand, a dynamic finite element for the same problem is also carried out by applying the ANSYS finite-element analysis system. Comparing the theoretical solution with finite-element solution, it can be found that two kinds of results obtained by making use of two different solving methods are suitably approached. Therefore, it is further concluded that the methods and computing processes of the theoretical solution are effective and accurate.     

模拟短脉冲激光在介质中传输的扩散综合有限元法

建立有限元模型,通过求解瞬态辐射传输方程模拟短脉冲激光在半透明介质中的传输.针对散射占优性半透明介质内辐射传输求解效率较差的问题,采用扩散综合加速迭代算法,提高计算效率,缩短计算时间.结果表明:采用精确解析式描述脉冲激光散射源项的求解策略可以获得准确的计算结果,精确地模拟快速变化的波前,不会产生数值扩散和数值振荡.此外,扩散综合迭代算法的计算时间仅为源项迭代的50%~60%.     

A solution of a non-homogeneous orthotropic cylindrical shell for axisymmetric plane strain dynamic thermoelastic problems

A solution of a non-homogeneous orthotropic elastic cylindrical shell for axisymmetric plane strain dynamic thermoelastic problems is developed. Firstly, a new dependent variable is introduced to rewrite the governing equation, the boundary conditions as well as the initial conditions. Secondly, a special function is introduced to transform the inhomogeneous boundary conditions to the homogeneous ones. Then by virtue of the orthogonal expansion technique, the equation with respect to the time variable is derived, of which the solution can be obtained. The displacement solution is finally presented, which can degenerate in a rather straightforward way to the solution for a homogeneous orthotropic cylindrical shell and isotropic solid cylinder as well as that for a non-homogeneous isotropic cylindrical shell. Using the present method, integral transform can be avoided. It is fit for a cylindrical shell with arbitrary thickness subjected to arbitrary thermal loads. It is also very convenient to deal with dynamic thermoelastic problems for different boundary conditions. Besides, the numerical calculation involved is very easy to be performed. Several examples are presented.     

Accurate numerical resolution of transients in initial-boundary value problems for the heat equation

If the initial and boundary data for a PDE do not obey an infinite set of compatibility conditions, singularities will arise in the solution at the corners of the initial time–space domain. For dissipative equations, such as the 1-D heat equation or 1-D convection–diffusion equations, the impacts of these singularities are short lived. However, they can cause a very severe loss of numerical accuracy if we are interested in transient solutions. The phenomenon has been described earlier from a theoretical standpoint. Here, we illustrate it graphically and present a simple remedy which, with only little extra cost and effort, restores full numerical accuracy.     

Orienting molecules using half-cycle pulses

Using a rigid-rotor model, we study the orientation dynamics of polar diatomic molecules excited by experimentally available half-cycle pulses. The results of the numerical solution of the time-dependent Schr?dinger equation are compared to those of an approximate “sudden-impact” impulsive model neglecting the molecular rotation during the pulse. We show that efficient orientation is achieved during time periods of several picoseconds for LiCl. For short pulses, where the kicked molecule model is valid, orientation turns out to be mainly sensitive to the time-integrated field amplitude and not the shape or rise time of the pulse. Received 16 August 2000 and Received in final form 4 December 2000     

皮秒类提升脉冲串的产生与传输

通过函数和变量变换给出变系数非线性薛定谔方程的一个新的精确解,它可以用来描述非均匀光纤中的调制不稳定过程.基于这个精确解我们给出一类新的类提升脉冲串,并讨论该脉冲串的传输特性.结果表明,当类提升脉冲串能量大于某一阈值时类提升脉冲串可以稳定传输.同时我们还比较指数控制系统和理想系统中类提升脉冲串的传输,结果表明选择合适的控制系统可以抑制脉冲间的相互作用,这对增加光通信的传输比特率是非常有益的.     

Early transient responses of multi-span stepped single-walled carbon nanotubes using the method of reverberation ray matrix

The early transient responses of multi-span stepped single walled carbon nanotubes (SWCNTs) under impact loadings are studied by the method of reverberation ray matrix (MRRM). The dynamics model of the carbon nanotubes is established in the Fourier phase space on the basis of the nonlocal Timoshenko beam model. The wave solutions of SWCNTs with arbitrary boundary conditions are obtained by the wave method. The reverberation ray matrix of the multi-span stepped SWCNTs is the product of scattering, phase and permutation matrices, which can be determined by the impact loadings, continuous conditions and boundary conditions. The early transient responses can be calculated by the inverse Fourier transform of the sum of initial ray groups. It can be found that the early transient displacement response in the very short time subjected to the impact loading is very small, while the transient transverse shear strain becomes large in the very short time. The influences of nanotubes span number, nanotubes type and boundary conditions on the early transient responses of multi-span stepped SWCNTs are investigated.     

三能级原子与少周期脉冲串作用中的相干布居捕获

研究了少周期脉冲串作用下三能级原子中的布居转移和相干布居捕获现象.在非旋波近似的情况下求解了密度矩阵方程.研究结果表明在等时间间隔的锁相脉冲作用下,系统能级的布居逐步转移并积累,系统基态相干也逐步积累.在满足脉冲重复频率为基态能级频差的整数分之一倍时,三能级系统和频率梳中两梳齿频率成分作用形成相干布居捕获现象,原子暗态布居值达到最大,介质对脉冲透明.在适当选取少周期脉冲参量的情况下,在0.5个ns的时间内三能级系统相干性演化到最大后到达稳态,相干布居捕获发生.与脉宽为100个fs的多周期脉冲相比,少周期脉冲串在介质中建立相干布居捕获的时间缩短两个数量级.由于频率梳中与三能级系统发生作用的梳频成份有相同的频移,相干布居捕获的条件双光子共振仍然满足.因而,当两基态能级频率差较大时,如果选取少周期脉冲载波频率为系统能级1至2和1至3的传输频率之和的一半ω=(ω₁+ω₂)/2,室温下原子热运动的引起的多普勒频移并不会破坏相干布居捕获.     

Temperature determination in the frozen region during cryosurgery of rabbit liver using MR image analysis

Cryosurgery currently is being used clinically to treat tumors in internal organs such as the liver and prostate. Although performed at present under ultrasound monitoring, magnetic resonance imaging (MRI)-guidance of these procedures not only permits monitoring of the frozen region during cryosurgery but also makes it possible to determine the temperature distribution in the frozen region, which is not possible using ultrasound monitoring. A good estimate of the region of destruction in the tissue can be obtained from correlating the temperature distribution and the time course of the freezing with the image of the frozen region. Unfortunately, MR pulse sequence-based temperature determination techniques such as diffusion, relazation time, and chemical shift cannot be used for measuring the temperature in the frozen region because the T₂ of the frozen region is so short that there is effectively no RF signal from the frozen region. This paper describes a numerical technique for determining the two dimensional temperature distribution in the frozen region during MR image-guided cryosurgery of normal liver in rabbits. The technique involves solving the energy equation numerically in the frozen region to determine the temperature distribution there. The boundary conditions needed to solve the equation are determined from MR images of the frozen tissue during cryosurgery and from the measured temperature of the cryoprobe. The calculated temperature in the frozen region is then correlated with the damaged region (cryolesion) determined from post mortem histologic evaluation.     

在双色场中使用相干叠加态来提高高次谐波的发射

本文提出了一种在双色场中使用相干叠加态来提高高次谐波发射效率的方法.我们首先通过在800 nm基本激光脉冲上添加一束控制激光脉冲在理论上得到双色合成激光场;其次,通过求解一维含时薛定谔方程,计算得到了体系处于不同初始态时的高次谐波谱,并结合时频分布图时其微观机理进行分析.计算结果表明,使用相干叠加态作为体系的初始态,通过调节两束激光的相对相位可以有效实现提高高次谐波的发射效率的目的.     

Simplified approach to the vibration analysis of elastic plates due to sonic boom

A theoretical study of the response of a flat plate to a sonic boom excitation is presented. For such a study, the problem of transient vibrations of elastic plates having clamped or simply supported boundary conditions under a pulse load in the shape of a capital N corresponding to a typical far-field sonic boom disturbance is discussed in a new fashion by using the concept of iso-amplitude contour lines on the surface of the plate. Series solutions consisting of products of eigenfunctions times functions of time are employed to analyse the motions. As an illustration of the technique, an elliptical plate subjected to a typical N wave arriving at normal incidence is chosen as a model because this involves a curvilinear boundary of a relatively simple shape, yet has no simple exact solution. Closed form solutions are obtained for both clamped as well as simply supported edges. The results have technical importance for the prediction of response of window panes and wall-panels to sonic boom. All details are explained by graphs.     

在双色场中使用相干叠加态来提高高次谐波的发射

本文提出了一种在双色场中使用相干叠加态来提高高次谐波发射效率的方法. 我们首先通过在800nm基本激光脉冲上添加一束控制激光脉冲在理论上得到双色合成激光场;其次,通过求解一维含时薛定谔方程,计算得到了体系处于不同初始态时的高次谐波谱,并结合时频分布图对其微观机理进行分析. 计算结果表明,使用相干叠加态作为体系的初始态,通过调节两束激光的相对相位可以有效实现提高高次谐波的发射效率的目的.