深穿透粒子输运问题的自适应抽样方法

屏蔽计算中的深穿透问题一直是蒙特卡罗计算的一个难题,研究了一种发射点作为驿站的随机游动机制,推导了相应的自适应抽样方法。其主要优势在于,在蒙特卡罗方法求解粒子输运的同时,利用已经获得的信息,自适应地控制各次抽样数,不断完善计算进程。通过对碰撞点引进重要性函数,实现发射点作为驿站的重要性抽样,并结合自适应控制达到最佳抽样状态。数值结果表明:基于发射点作为驿站的自适应抽样方法,在一定程度上克服了深穿透计算中估计值偏低现象。相应的重要函数抽样方法获得了满意的结果。     

Improved method of measuring pressure coupled response for composite solid propellants

Pressure coupled response is one of the main causes of combustion instability in the solid rocket motor. It is also a characteristic parameter for predicting the stability. The pressure coupled response function is usually measured by different methods to evaluate the performance of new propellant. Based on T-burner and “burning surface doubled and secondary attenuation”, an improved method for measuring the pressure coupled response of composite propellant is introduced in this article. A computational fluid dynamics (CFD) study has also been conducted to validate the method and to understand the pressure oscillation phenomenon in T-burner. Three rounds of tests were carried out on the same batch of aluminized AP/HTPB composite solid propellant. The experimental results show that the sample propellant had a high response function under the conditions of high pressure (~11.5 MPa) and low frequency (~140 Hz). The numerically predicted oscillation frequency and amplitude are consistent with the experimental results. One practical solid rocket motor using this sample propellant was found to experience pressure oscillation at the end of burning. This confirms that the sample propellant is prone to combustion instability. Finally, acoustic pressure distribution and phase difference in T-burner were analyzed. Both the experimental and numerical results are found to be associated with similar acoustic pressure distribution. And the phase difference analysis showed that the pressure oscillations at the head end of the T-burner are 180° out of phase from those in the aft end of the T-burner.     

Irreversible Monte Carlo algorithms for efficient sampling

Equilibrium systems evolve according to Detailed Balance (DB). This principle guided the development of Monte Carlo sampling techniques, of which the Metropolis–Hastings (MH) algorithm is the famous representative. It is also known that DB is sufficient but not necessary. We construct irreversible deformation of a given reversible algorithm capable of dramatic improvement of sampling from known distribution. Our transformation modifies transition rates keeping the structure of transitions intact. To illustrate the general scheme we design an Irreversible version of Metropolis–Hastings (IMH) and test it on an example of a spin cluster. Standard MH for the model suffers from critical slowdown, while IMH is free from critical slowdown.     

耦合不可压流场输运方程的格子Boltzmann方法研究

基于格子Boltzmann方法,提出了求解耦合不可压缩流场输运方程的一种改进数值方法. 该方法使用格子Boltzmann方法求解流场方程,并根据流场格子模型的密度分布函数构建了输运方程的二阶离散格式. 通过二维平板通道流场输运系统验证了该方法的有效性.数值结果表明,该方法可以有效地减少计算过程中出现的非物理耗散, 并克服了传统模型所需巨大存储量的缺点.     

Reciprocal relations for nonlinear coupled transport

Reciprocal relations are derived for transport between two pairs of reservoirs energetically coupled at mesoscopic contact points through second order in the chemical potential differences. In the derivation attention is focussed on the average number of particles transported between the reservoirs during each excursion of the contact point away from, and regression back to, a steady state. All quantities involved are experimentally accessible in the full counting statistics of the transport processes.     

Multicomponent transport algorithms for partially ionized mixtures

We investigate iterative methods for solving linear systems arising from the kinetic theory of gases and providing multicomponent transport coefficients of partially ionized plasmas. We consider the situations of weak and strong magnetic fields as well as electron temperature nonequilibrium and the linear systems are investigated in their natural constrained singular symmetric form. Stationary iterative techniques are considered with new more singular formulations of the transport linear systems as well as orthogonal residuals algorithms. The new formulations are derived by considering generalized inverses with nullspaces of increasing dimension. Numerical tests are performed with high temperature air and iterative techniques lead to fast and accurate evaluation of the transport coefficients for all ionization levels and magnetic field intensities.     

Improved Lanczos algorithms for blackbox MRS data quantitation

Magnetic resonance spectroscopy (MRS) has been shown to be a potentially important medical diagnostic tool. The success of MRS depends on the quantitative data analysis, i.e., the interpretation of the signal in terms of relevant physical parameters, such as frequencies, decay constants, and amplitudes. A variety of time-domain algorithms to extract parameters have been developed. On the one hand, there are so-called blackbox methods. Minimal user interaction and limited incorporation of prior knowledge are inherent to this type of method. On the other hand, interactive methods exist that are iterative, require user involvement, and allow inclusion of prior knowledge. We focus on blackbox methods. The computationally most intensive part of these blackbox methods is the computation of the singular value decomposition (SVD) of a Hankel matrix. Our goal is to reduce the needed computational time without affecting the accuracy of the parameters of interest. To this end, algorithms based on the Lanczos method are suitable because the main computation at each step, a matrix-vector product, can be efficiently performed by means of the fast Fourier transform exploiting the structure of the involved matrix. We compare the performance in terms of accuracy and efficiency of four algorithms: the classical SVD algorithm based on the QR decomposition, the Lanczos algorithm, the Lanczos algorithm with partial reorthogonalization, and the implicitly restarted Lanczos algorithm. Extensive simulation studies show that the latter two algorithms perform best.     

改进的3维粒子模拟并行算法

 提出了两种改进的3维粒子模拟并行算法,改进的并行算法能在每个时间步减少一次进程同步。算法分析和数值模拟表明,由于粒子运动路径和发射的初始位置与随机函数有关,只有一种改进的并行算法能保证并行计算正确。在3维粒子模拟软件CHIPIC3D上实现了改进的并行算法,应用CHIPIC3D对一种相对论返波管进行了并行模拟,模拟结果表明改进的并行算法能取得更高的加速比和效率。     

改进的3维粒子模拟并行算法

提出了两种改进的3维粒子模拟并行算法,改进的并行算法能在每个时间步减少一次进程同步。算法分析和数值模拟表明,由于粒子运动路径和发射的初始位置与随机函数有关,只有一种改进的并行算法能保证并行计算正确。在3维粒子模拟软件CHIPIC3D上实现了改进的并行算法,应用CHIPIC3D对一种相对论返波管进行了并行模拟,模拟结果表明改进的并行算法能取得更高的加速比和效率。     

Application of the ICE-PIC method for the dimension reduction of chemical kinetics coupled with transport

A new dimension-reduction method, the Invariant Constrained-equilibrium Edge Pre-Image Curve (ICE-PIC) method, to simplify chemical kinetics has recently been developed by Ren et al. [Z. Ren, S.B. Pope, A. Vladimirsky, J.M. Guckenheimer, J. Chem. Phys. 124 (2006) 114111]. In the present work, the ICE-PIC method is first applied to the homogeneous autoignition of stoichiometric methane/air and its accuracy is shown to compare favorably to those of other methods (QSSA and RCCE). For inhomogeneous systems such as flames, spatial transport by molecular diffusion causes a small perturbation of the composition away from the attracting, low-dimensional, invariant manifold identified by the ICE-PIC method. A “close-parallel” assumption is introduced which allows this perturbation to be determined, and leads to an additional “transport coupling” term in the evolution equation for the reduced variables. For the test case of a steady, one-dimensional, laminar, methane/air flame, it is shown that the inclusion of transport coupling can reduce the dimension-reduction errors by a factor of 100. The ICE-PIC method with eight degrees of freedom (including transport coupling) exhibits comparable accuracy to a quasi-steady state assumption (QSSA) reduced mechanism with 12 degrees of freedom.     

非定常辐射输运问题的蒙特卡罗自适应偏倚抽样

用蒙特卡罗(MC)方法模拟高温、高压、多介质、大变形辐射输运问题时,由于网格体积悬殊,导致各网格通量的统计误差涨落很大,随着时间步的增加,误差积累甚至会导致计算结果失真.为此,发展了针对全局网格计算的源偏倚抽样技巧.用于源偏倚抽样的价值函数基于上个时间步各网格通量及误差,通过加权构造产生,它比传统MC通过解伴随方程获取价值的性价比要高得多.数值试验表明,全局源偏倚抽样通过自适应分配当前时间步各网格的粒子数,有效地降低了当前步重要网格通量误差. 关键词： 非定常 辐射输运 蒙特卡罗 源偏倚抽样     

Static and dynamic routing and wavelength assignment algorithms for future transport networks

In this paper, two static and three dynamic routing algorithms have been proposed and compared to some of the existing algorithms on the basis of blocking probability. The two proposed static routing and wavelength assignment (RWA) algorithms reduce the blocking probability to maximize the utilization of network. For dynamic algorithms, first a model with no weights assignments is presented and then three algorithms are proposed and analysed with weight assignment resulting in reduction of blocking probability. All these algorithms are analysed and compared with four wavelength assignment schemes which are first-fit, random, most used and least used. It is shown that our proposed static algorithms give the best performance for first-fit wavelength assignment and most used wavelength assignment strategies with reduced complexity. For least used wavelength assignment and random wavelength assignment, 1 fixed and 2 alternate routing algorithm gives the lowest blocking probability. For dynamic routing, it has been shown that our proposed algorithm “less weight to maximum empty and nearest” gives the least blocking probability as compared to the other dynamic routing algorithms for random, most used and least used wavelength assignment strategies.     

改进的遥感图像多种信息熵阈值分割算法

给出了一种新的图像信息量的表述方法,提出了五种基于信息熵的科学分割遥感图像的改进方法。用MATLAB对这些算法进行了实验仿真,并进行了算法性能优化。得出了各种不同算法的最佳阈值和耗时,并对它们进行了比较和分析。实验数据表明,新算法较以往的算法在分割效果和速度上都有明显的改善和提高。     

Improved expression for calculating electron transport cross sections

We report a simplified correction for the electron transport cross sections (TCSs) for a number of selected atomic targets ranging from H to U and electron energies between 50 and 4000 eV. The correction has been made to the approximate analytical expression of transport cross sections derived by Jablonski [A. Jablonski, Phys. Rev. B 58 (1998) 16470] where an argued parameter is introduced. The latter is obtained from a polynomial fit. The energy dependence of the percentage deviation between TCSs from the corrected expression and those obtained from other sources is presented. The TCSs calculated in the present work showed better agreement with accurate values of TCSs than those reported in earlier publications. This may facilitate the evaluation of parameters needed for quantitative Auger-electron spectroscopy and X-ray photoelectron spectroscopy.     

二维耦合定向输运模型研究

建立了外部驱动力及噪声作用下的二维耦合定向输运模型, 其中的一个维度上为周期性分段棘齿势, 另一垂直维度上为周期性对称非棘齿势, 外部驱动力及噪声加在周期对称非棘齿势方向上, 而棘齿势方向不加任何驱动, 采用非平衡统计及非线性动力学理论研究了过阻尼情况下耦合系统在两个维度上的输运性质. 结果显示, 棘齿势与非棘齿势方向均可产生定向输运, 其中棘齿势方向的系统平均速度对耦合强度、噪声强度、驱动力强度及粒子数目均有明显的依赖性, 合适的耦合强度、噪声强度、驱动力强度或粒子数目下均可产生最大输运速度. 而非棘齿势方向的系统平均速度受非棘齿势势垒高度影响显著, 但随耦合强度、驱动力强度、驱动力初相位差及粒子数目的变化均出现波动现象, 表现出平均速度对这些参量的依赖性较弱.     

Spin-polarized transport through a coupled double-dot

We investigate the quantum transport through a mesoscopic device consisting of an open, lateral double-quantum-dot coupled by time oscillating and spin-polarization dependent tunneling which results from a static magnetic field applied in the tunneling junction. In the presence of a non-vanishing bias voltage applied to two attached macroscopic leads both spin and charge currents are driven through the device. We demonstrate that the spin and charge currents are controllable by adjusting the gate voltage, the frequency of driving field and the magnitude of the magnetic field as well. An interesting resonance phenomenon is observed.     

周期耦合量子阱中的输运问题

采用Gurvitz等人直接求解薛定谔方程的方法并结合数值计算,分析了驱动频率对周期耦合量子阱体系的电流的影响.结果表明：当驱动频率小于耦合量子阱间的能级差时,随着驱动频率的增大,系统平衡时的电流增加,当驱动频率大于耦合量子阱间能级差时,随着驱动频率的增大,平衡时的电流减小.这样,通过控制外场驱动频率来达到控制电流的目的. 关键词： 量子阱 驱动频率 电流     

Improved design for SOI based evanescently coupled multilayer spot-size converter

We report an improved version of a spot-size converter (SSC) consisting of a silicon nanowire evanescently coupled to a phase-matched Poly-Si multilayer structure. With wider transversal dimensions the multilayer structure expands the mode significantly thus increasing the coupling efficiency with the conventional single-mode fiber. Detailed optimization process of a 17-layer based SSC is discussed and its coupling efficiency with a high-NA fiber of radius 2 μm is obtained as 98% providing only 0.087 dB loss. Vertical alignment tolerance between the optimized SSC and a high-NA fiber of radius 2 μm is also shown. This novel design does not consist of a taper and can be fabricated by using CMOS compatible process. It has a short device length and more relaxed alignment tolerances with the fiber. Full-vectorial and computationally efficient finite element method and the least squares boundary residual method have been used for the analysis and optimization of the proposed structure.