Burton-Miller改进型边界方程的多频计算方法

提出了一种采用Burton-Miller改进型边界积分方程进行多频计算的方法。将Burton-Miller方程中的高奇异积分转化为弱奇异积分形式,获得Burton-Miller改进型边界积分方程;将方程中格林函数进行Taylor级数展开,并把波数从方程中分离出来,从而使随波数变化的计算矩阵表示为波数的矩阵级数形式。数值分析表明,本方法不仅保证了解在全波数范围内的唯一性,并且计算频率点数较多时可以节约大量时间,提高计算效率。      

A unified approach for predicting sound radiation from baffled rectangular plates with arbitrary boundary conditions

This paper discusses sound radiation from a baffled rectangular plate with each of its edges arbitrarily supported in the form of elastic restraints. The plate displacement function is universally expressed as a 2-D Fourier cosine series supplemented by several 1-D series. The unknown Fourier expansion coefficients are then determined by using the Rayleigh-Ritz procedure. Once the vibration field is solved, the displacement function is further simplified to a single standard 2-D Fourier cosine series in the subsequent acoustic analysis. Thus, the sound radiation from a rectangular plate can always be obtained from the radiation resistance matrix for an invariant set of cosine functions, regardless of its actual dimensions and boundary conditions. Further, this radiation resistance matrix, unlike the traditional ones for modal functions, only needs to be calculated once for all plates with the same aspect ratio. In order to determine the radiation resistance matrix effectively, an analytical formula is derived in the form of a power series of the non-dimensional acoustic wavenumber; the formula is mathematically valid and accurate for any wavenumber. Several numerical examples are presented to validate the formulations and show the effect of the boundary conditions on the radiation behavior of planar sources.     

New Way in Few-Body Scattering Calculations

The key features of the new general approach to solution of few-body scattering problems in hadronic, nuclear and atomic physics are presented and discussed in the paper. The approach is based on a general idea of the lattice-like discretization of few-body continuum using the stationary wave-packet basis in momentum space. The new technique includes an efficient averaging and smoothing of singular kernels of the scattering integral equations over the lattice cells. So, such an averaging procedure allows us to transform the complicated singular integral kernels into usual matrices with regular and smooth matrix elements. Such a transformation is shown to lead to an enormous simplification of the solving procedure for scattering equations.     

Generalized Multivariate Singular Spectrum Analysis for Nonlinear Time Series De-Noising and Prediction

Singular spectrum analysis and its multivariate or multichannel singular spectrum analysis(MSSA)variant are effective methods for time series representation,denoising and prediction,with broad application in many fields.However,a key element in MSSA is singular value decomposition of a high-dimensional matrix stack of component matrices,where the spatial(structural)information among multivariate time series is lost or distorted.This vector-space model also leads to difficulties including high dimensionality,small sample size,and numerical instability when applied to multi-dimensional time series.We present a generalized multivariate singular spectrum analysis(GMSSA)method to simultaneously decompose multivariate time series into constituent components,which can overcome the limitations of conventional multivariate singular spectrum analysis.In addition,we propose a Samp En-based method to determine the dominant components in GMSSA.We demonstrate the effectiveness and efficiency of GMSSA to simultaneously de-noise multivariate time series for attractor reconstruction,and to predict both simulated and real-world multivariate noisy time series.     

A NUMERICAL METHOD FOR SCATTERING FROM ACOUSTICALLY SOFT AND HARD THIN BODIES IN TWO DIMENSIONS

This paper presents a numerical method for predicting the acoustic scattering from two-dimensional (2-D) thin bodies. Both the Dirichlet and Neumann problems are considered. Applying the thin-body formulation leads to the boundary integral equations involving weakly singular and hypersingular kernels. Completely regularizing these kinds of singular kernels is thus the main concern of this paper. The basic subtraction-addition technique is adopted. The purpose of incorporating a parametric representation of the boundary surface with the integral equations is two-fold. The first is to facilitate the numerical implementation for arbitrarily shaped bodies. The second one is to facilitate the expansion of the unknown function into a series of Chebyshev polynomials. Some of the resultant integrals are evaluated by using the Gauss-Chebyshev integration rules after moving the series coefficients to the outside of the integral sign; others are evaluated exactly, including the modified hypersingular integral. The numerical implementation basically includes only two parts, one for evaluating the ordinary integrals and the other for solving a system of algebraic equations. Thus, the current method is highly efficient and accurate because these two solution procedures are easy and straightforward. Numerical calculations consist of the acoustic scattering by flat and curved plates. Comparisons with analytical solutions for flat plates are made.     

Essential variables in the integrability problem of planar vector fields

This Letter is devoted to the integrability problem of planar nonlinear differential equations. We introduce a new method to detect local analytic integrability or to construct a singular series expansion of the first integral around a singular point for planar vector fields. The method allows to find new variables (essential variables) where the integrability problem is more feasible. The new method can be used in different context and is an alternative to all the methods developed up to now for any particular case.     

Theoretical and computational aspects of scattering from periodic surfaces: one-dimensional transmission interface

We consider the scattering from and transmission through a one-dimensional periodic surface. For this problem, the electromagnetic cases of TE and TM polarization reduce to the scalar acoustic examples. Three different theoretical and computational methods are described, all involving the solution of integral equations and their resulting discrete matrix system of equations for the boundary unknowns. They are characterized by two sample spaces for their discrete solution, coordinate space and spectral space, and labelled by the sampling of the rows and columns of the discretized matrices. They are coordinate-coordinate (CC), the usual coordinate-space method, spectral-coordinate (SC) where the matrix rows are discretized or sampled in spectral space and spectral-spectral (SS) where both rows and columns are sampled in spectral space. The SS method uses a new topological basis expansion for the boundary unknowns.

Equations are derived for infinite surfaces, then specialized and solved for periodic surfaces. Computational results are presented for the transmission problem as a function of roughness, near-grazing incidence as well as many other angles, density and wavenumber ratios. Matrix condition numbers and different sampling methods are considered. An error criterion is used to gauge the validity of the results.

The computational results indicated that the SC method was by far the fastest (by several orders of magnitude), but that it became ill-conditioned for very rough surfaces. The CC method was most reliable, but often required very large matrices and was consequently extremely slow. It is shown that the SS method is computationally efficient and accurate at near-grazing incidence and can be used to fill a gap in the literature. Extensive computational results indicate that both SC and SS are highly robust computational methods. Spectral-based methods thus provide viable computational schemes to study periodic surface scattering.     

A Nyström method for weakly singular integral operators on surfaces

We describe a modified Nyström method for the discretization of the weakly singular boundary integral operators which arise from the formulation of linear elliptic boundary value problems as integral equations. Standard Nyström and collocation schemes proceed by representing functions via their values at a collection of quadrature nodes. Our method uses appropriately scaled function values in lieu of such representations. This results in a scheme which is mathematically equivalent to Galerkin discretization in that the resulting matrices are related to those obtained by Galerkin methods via conjugation with well-conditioned matrices, but which avoids the evaluation of double integrals. Moreover, we incorporate a new mechanism for approximating the singular integrals which arise from the discretization of weakly singular integral operators which is considerably more efficient than standard methods. We illustrate the performance of our method with numerical experiments.     

Theoretical and computational aspects of scattering from periodic surfaces: one-dimensional transmission interface

Abstract

We consider the scattering from and transmission through a one-dimensional periodic surface. For this problem, the electromagnetic cases of TE and TM polarization reduce to the scalar acoustic examples. Three different theoretical and computational methods are described, all involving the solution of integral equations and their resulting discrete matrix system of equations for the boundary unknowns. They are characterized by two sample spaces for their discrete solution, coordinate space and spectral space, and labelled by the sampling of the rows and columns of the discretized matrices. They are coordinate-coordinate (CC), the usual coordinate-space method, spectral-coordinate (SC) where the matrix rows are discretized or sampled in spectral space and spectral-spectral (SS) where both rows and columns are sampled in spectral space. The SS method uses a new topological basis expansion for the boundary unknowns.

Equations are derived for infinite surfaces, then specialized and solved for periodic surfaces. Computational results are presented for the transmission problem as a function of roughness, near-grazing incidence as well as many other angles, density and wavenumber ratios. Matrix condition numbers and different sampling methods are considered. An error criterion is used to gauge the validity of the results.

The computational results indicated that the SC method was by far the fastest (by several orders of magnitude), but that it became ill-conditioned for very rough surfaces. The CC method was most reliable, but often required very large matrices and was consequently extremely slow. It is shown that the SS method is computationally efficient and accurate at near-grazing incidence and can be used to fill a gap in the literature. Extensive computational results indicate that both SC and SS are highly robust computational methods. Spectral-based methods thus provide viable computational schemes to study periodic surface scattering.     

High-order solutions of three-dimensional rough-surface scattering problems at high-frequencies. Part II: The vector electromagnetic case

We introduce a new numerical scheme for three-dimensional electromagnetic rough-surface scattering simulations with the capability of delivering very accurate results from low to high frequencies at a cost that is independent of the wavelength of radiation. The method is an extension of the ideas and techniques introduced in the first paper of this series (Waves in Random and Complex Media, 15 (2005), pp. 1-16) to the vector electromagnetic case, and it is based on the solution of an integral equation formulation of the scattering problem. As in the scalar case, the solution of the integral equation (i.e. the current) is expressed as a slow modulation of an oscillatory exponential of known phase, and explicit recursive formulae are derived for the successive terms in a series expansion of the slow envelope in inverse powers of the wavenumber. As we show, and in spite of the considerably more involved nature of the derivations and resulting formulae, the performance of the method retains the favourable characteristics that were demonstrated in the treatment of acoustic scattering problems. In particular, results with full double-precision accuracy are presented for various geometries, incidences and polarizations.     

Calculation of the collision integral linear kernel for Maxwellian molecules in the isotropic case

Application of the method of nonlinear moments to solve the Boltzmann equation generates the need to sum a series that is the expansion of the distribution function in basis functions. This series converged only if the Grad test is fulfilled. Such a limitation can be removed if the expansion of the distribution function is summed over the index related to only the expansion in velocity magnitude. In this case, the distribution function and the collision integral become expanded in only spherical harmonics and the expansion coefficients satisfy integro-differential equations. The kernels of these equations are the sums of the Sonine polynomials in the velocities of colliding and outgoing particles multiplied by matrix elements of the collision integral. For a number of arguments, the direct calculation of the kernels requires that a very large number of terms in the sum be taken into consideration. In this respect, an approach seems to be promising in which the asymptotics of the matrix elements and Sonine polynomials at large indices are used and summation over index is replaced by integration. In this paper, we apply this approach to calculate the linear kernel in the isotropic case, assuming that interaction between particles is described by a pseudopower law. With this approach, the collision integral kernel can be calculated with a high accuracy using as little as a few tens of series terms and the asymptotic estimate of the residue.     

Power Series Expansion of Propagator for Path Integral and Its Applications

In this paper we obtain a propagator of path integral for a harmonic oscillator and a driven harmonic oscillator by using the power series expansion. It is shown that our result for the harmonic oscillator is more exact than the previous one obtained with other approximation methods. By using the same method, we obtain a propagator of path integral for the driven harmonic oscillator, which does not have any exact expansion. The more exact propagators may improve the path integral results for these systems.     

基于奇异值分解和中位数绝对偏差的拉曼成像数据去噪方法

拉曼成像是一种无损伤、无需标记的光谱成像技术,它可以提供样品的不同组分的分子指纹信息以及空间分布特征,相比其他成像技术有着更重要的应用。但是拉曼散射的截面积小,灵敏度低,加上在很多实验中为了观察某些组分的动态分布而缩短扫描时间,导致最终得到的成像数据被噪声干扰,因此往往需要对信号进行去噪处理。常规的算法一般都是基于一个给定的数学模型对光谱进行处理,容易造成过滤波,使得信号失真;另外,在处理拉曼成像数据时,常规算法往往是对数据进行逐条光谱去噪,从而忽略了多条光谱之间的相互关系,导致最终的拉曼图像仍然受许多噪点干扰。因此,提出了一种基于奇异值分解和中位数绝对偏差的拉曼成像的信号处理方法,用于拉曼成像数据的去噪处理。该方法首先对拉曼成像数据进行奇异值分解,获得一个奇异值矩阵与两个正交矩阵;然后通过中位数绝对偏差法对奇异值矩阵中的各奇异值进行离群值检测,选取前k个被连续标记的离群值作为要保留的奇异值,并将其余的奇异值赋值为零,得到新的奇异值矩阵;最后用新的奇异值矩阵与两个正交矩阵重新求解得到去噪后的拉曼成像数据。实验中,首先验证了中位数绝对偏差法确定前k个奇异值的正确性,其次分别从处理后的图像质量和信号波形两方面对比了该算法与常规算法的去噪效果。结果证明,中位数绝对偏差法可以快速地确定出合理的k值大小,而且,依据该k值处理后的成像数据不仅在成像质量上消除了大量的噪点,使得组分的空间分布特征清晰可见,也在信号波形上较完美地保留了微小谱峰,并恢复光谱信号。该算法不同于常规算法,能同时对整个拉曼成像数据进行处理,并保留光谱之间的统计特征,是一种更加有效的拉曼成像数据的去噪方法。     

The quantification of structure-borne transmission paths by inverse methods. Part 2: Use of regularization techniques

The inversion of an ill-conditioned matrix of measured data lies at the heart of procedures for the quantification of structure-borne sources and transmission paths. In an earlier paper the use of over-determination, singular value decomposition and the rejection of small singular values was discussed. In the present paper alternative techniques for regularizing the matrix inversion are considered. Such techniques have been used in the field of digital image processing and more recently in relation to nearfield acoustic holography. The application to structure-borne sound transmission involves matrices, which vary much more with frequency and from one element to another. In this study Tikhonov regularization is used with the ordinary cross-validation method for selecting the regularization parameter. An iterative inversion technique is also studied. Here a form of cross-validation is developed allowing an optimum value of the iteration parameter to be selected. Simulations are carried out using a rectangular plate structure to assess the relative merits of these techniques. Experiments are also performed to validate the results. Both techniques are found to give considerably improved results compared to singular value rejection.     

INCOMPLETE q-GAMMA FUNCTION AND TRICOMI EXPANSION

In this paper, we introduce a q-analogue of the Tricomi expansion for the incomplete q-gamma function. A general method is described for converting a power series into an expansion in incomplete q-gamma function. Also, we use the q-Tricomi expansion for giving a formal proof of the relation between the incomplete gamma function and the exponential integral. Finally, we formally deduce the q-Tricomi expansion via the q-Taylor expansion.     

Response of infinite length rods and beams with periodically varying area

This paper develops a solution method for the longitudinal motion of a rod or the flexural motion of a beam of infinite length whose area varies periodically. The conventional rod or beam equation of motion is used with the area and moment of inertia expressed using analytical functions of the longitudinal (horizontal) spatial variable. The displacement field is written as a series expansion using a periodic form for the horizontal wavenumber. The area and moment of inertia expressions are each expanded into a Fourier series. These are inserted into the differential equations of motion and the resulting algebraic equations are orthogonalized to produce a matrix equation whose solution provides the unknown wave propagation coefficients, thus yielding the displacement of the system. An example problem of both a rod and beam are analyzed for three different geometrical shapes. The solutions to both problems are compared to results from finite element analysis for validation. Dispersion curves of the systems are shown graphically. Convergence of the series solutions is illustrated and discussed.     

Spectral Analysis of Unitary Band Matrices

 This paper is devoted to the spectral properties of a class of unitary operators with a matrix representation displaying a band structure. Such band matrices appear as monodromy operators in the study of certain quantum dynamical systems. These doubly infinite matrices essentially depend on an infinite sequence of phases which govern their spectral properties. We prove the spectrum is purely singular for random phases and purely absolutely continuous in case they provide the doubly infinite matrix with a periodic structure in the diagonal direction. We also study some properties of the singular spectrum of such matrices considered as infinite in one direction only. Received: 29 April 2002 / Accepted: 7 August 2002 Published online: 20 January 2003 Communicated by B. Simon     

High-order solutions of three-dimensional rough-surface scattering problems at high frequencies. I: the scalar case

We present a new high-order numerical method for the solution of high-frequency scattering problems from rough surfaces in three dimensions. The method is based on the asymptotic solution of appropriate integral equations in the high-frequency regime, in a manner that bypasses the need to resolve the fields on the scale of the wavelength of radiation. Indeed, inspired by prior work in two dimensions, we seek a solution of the integral equation in the form of a slow modulation of the incoming radiation, and we choose a series expansion in inverse powers of the wavenumber to represent the unknown slowly varying envelope. As we show, this framework can be made to yield an efficiently computable recursion for the terms in the series to any arbitrary order. The resulting algorithms generally provide a very significant improvement over classical (e.g. Kirchhoff) approximations in both accuracy and applicability and they can, in fact, effectively produce results with full double-precision accuracy for configurations of practical interest and up to the resonance regime.     

Sense-through-wall human detection using the UWB radar with sparse SVD

In this paper, a method for through-wall human detection based on the singular values decomposition of the measurement matrices is presented. After demonstrating the sparsity of the matrices using the CLEAN algorithm, an SVD algorithm based on the Lanczos process is applied to compute their singular values. We also analyze the singular values of matrices constructed by difference square techniques for different types of walls and compare our algorithm with a 2-D imaging approach proposed by researchers in Time Domain Company. Detection results show that our method performs well in the gypsum wall, brick wall, and wooden door.     

Domain-partition power series in vibration analysis of variable-cross-section rods

This paper presents a power series method with domain partition implemented in a matrix formulation, as an alternative to other power series techniques in vibration analysis. The proposed method solves linear differential equations efficiently up to a desired degree of accuracy and remedies two limitations of the conventional power series method. One limitation is related to the convergence domain of the series solution. If this domain does not include the region under analysis, the series expansion gives meaningless results. The other limitation is computational in nature; numerical difficulties arise when calculating natural frequencies, modes of vibration and dynamic stiffness of continuous models at high frequency. To compare some of the available implementations of the power series method in modal analysis, the longitudinal vibration of a rod with linearly varying area is studied. By means of this simple example, it is demonstrated that the power series method with domain partition provides more versatility than the power series approximation on complete domains.