Grid Restrained Nelder-Mead Algorithm

Probably the most popular algorithm for unconstrained minimization for problems of moderate dimension is the Nelder-Mead algorithm published in 1965. Despite its age only limited convergence results exist. Several counterexamples can be found in the literature for which the algorithm performs badly or even fails. A convergent variant derived from the original Nelder-Mead algorithm is presented. The proposed algorithm's convergence is based on the principle of grid restrainment and therefore does not require sufficient descent as the recent convergent variant proposed by Price, Coope, and Byatt. Convergence properties of the proposed grid-restrained algorithm are analysed. Results of numerical testing are also included and compared to the results of the algorithm proposed by Price et al. The results clearly demonstrate that the proposed grid-restrained algorithm is an efficient direct search method.     

Sticking efficiency of nanoparticles in high-velocity collisions with various target materials

In order to find reliable collector surfaces for the Mesospheric Aerosol – Genesis, Interaction and Composition (MAGIC) sounding rocket experiment, intended to collect atmospheric nanoparticles, the sticking efficiency of nanoparticles was measured on several targets of different materials. The nanoparticles were generated by a molecular beam apparatus in Jena, Germany, by laser ablation (Al₂O₃ particles, diameter 5–50 nm) and by laser pyrolysis (carbon particles, diameter 10–20 nm). In a vacuum environment (>10⁻⁵ mbar) the particles condensed from the gas phase, formed a particle beam, and were accelerated to ∼ ∼1 km/s. The sticking efficiency on the target materials carbon, gold and grease was measured by a microbalance. Results demonstrate moderate to high sticking probabilities. Thus, the capture and retrieval of atmospheric nanoparticles was found to be quantitatively feasible.     

Multigrid convergence of inviscid fixed‐ and rotary‐wing flows

The affect of multigrid acceleration implemented within an upwind‐biased Euler method is presented, and applied to fixed‐wing and rotary‐wing flows. The convergence of fixed‐ and rotary‐wing computations is shown to be vastly different, and multigrid is shown to be less effective for rotary‐wing flows. The flow about a hovering rotor suffers from very slow convergence of the inner blade region, where the flow is effectively incompressible. Furthermore, the vortical wake must develop over several turns before convergence is achieved, whereas for fixed‐wing computations the far‐field grid and solution have little significance. Results are presented for single mesh and two, three, four, and five level multigrid, and using five levels a reduction in required CPU time of over 80 per cent is demonstrated for rotary‐wing computations, but 94 per cent for fixed‐wing computations. It is found that a simple V‐cycle is the most effective, smoothing in the decreasing mesh density direction only, with a relaxed trilinear prolongation operator. Copyright © 2002 John Wiley & Sons, Ltd.     

基于Agent的教育网格体系结构

针对教育资源信息量庞大、种类多、表现形式多样、存储分散等特点．提出一种基于Agent的5层教育网格体系结构，定义了层次间不同类别网格中间件．在本结构模型中引进Agent技术实现层次间不同类别中间件的构造，可获取并重构满足用户特定需求的教育资源构件，达到自主性、个性化组织教学内容之目的．     

Temperature and deformation measurements in transient metal cutting

Metal cutting is a thermomechanically coupled process in which plasticity induced heating and friction play a critical role. The objective of this work is to develop a methodology to understand and quantify this coupling. Temperatures of the workpiece and the chip during transient cutting processes are measured using a linear array of 16 InSb infrared detectors with 200 ns rise time and 27 μm spatial resolution. Three different materials, 1018 CR steel, Al6061-T6 and Ti-6Al-4V, are tested at a cutting speed of 4.3 m s⁻¹. A grid method is used to measure deformations during the above set of experiments. Measured values of temperature and deformation are compared to results of finite element simulations of the experiments.     

An accurate moving grid Eulerian Lagrangian localized adjoint method for solving the one‐dimensional variable‐coefficient ADE

An accurate finite‐volume Eulerian Lagrangian localized adjoint method (ELLAM) is presented for solving the one‐dimensional variable coefficients advection dispersion equation that governs transport of solute in porous medium. The method uses a moving grid to define the solution and test functions. Consequently, the need for spatial interpolation, or equivalently numerical integration, which is a major issue in conventional ELLAM formulations, is avoided. After reviewing the one‐dimensional method of ELLAM, we present our strategy and detailed calculations for both saturated and unsaturated porous medium. Numerical results for a constant‐coefficient problem and a variable‐coefficient problem are very close to analytical and fine‐grid solutions, respectively. The strength of the developed method is shown for a large range of CFL and grid Peclet numbers. Copyright 2004 John Wiley & Sons, Ltd.     

多块网格中的块隐式方法

本文实现了逐点扫描的块隐式算法，并根据其特点，提出了一种基于该算法的多块网格方法．通过对几个具有基准解问题的数值计算发现：该算法相对于ＳＩＭＰＬＥ算法而言，其收敛速度较快；本文所提出的多块网格方法是可行的．     

网格紧致化存储系统的分区设计和ABC分类分析

基于网格的紧致化存储系统可以密集地存储货物,在充分利用存储空间的同时还能较好地保证货物的可达性。现有研究集中在取货时间最小化算法设计等基本问题上,但如何更好地利用这些研究结果改善系统设计的研究还未展开。本文应用现有算法研究成果分析网格紧致化存储系统的不同分区布局的绩效,构建了一个开放的排队网络,由基于取货时间的分区仓储子系统、仓储分区内基于货物需求频率的ABC分类子系统两个部分组成。以网格面积10×10为实例的计算实验表明,与随机存储分配规则相比,对基于网格的存储系统进行分区所实现的期望取货时间可以减少53.8%;在此基础上再进行ABC分类可以进一步改善系统性能最多可以达到46.5%,同时还可以大幅度减少因排队所产生的等待时间。     

A new two grid variational multiscale method for steady‐state natural convection problem

A two‐grid variational multiscale method based on two local Gauss integrations for solving the stationary natural convection problem is presented in this article. A significant feature of the method is that we solve the natural convection problem on a coarse mesh using finite element variational multiscale method based on two local Gauss integrations firstly, and then find a fine grid solution by solving a linearized problem on a fine grid. In the computation, we introduce two local Gauss integrations as a stabilizing term to replace the projection operator without adding other variables. The stability estimates and convergence analysis of the new method are derived. Ample numerical experiments are performed to validate the theoretical predictions and demonstrate the efficiency of the new method. Copyright © 2016 John Wiley & Sons, Ltd.     

基于非结构自适应网格的二维Euler方程数值求解方法研究

提出了一种基于非结构自适应网格的二维Euler方程的数值解法.采用有限体积法进行空间离散,通量计算采用Jamson中心格式,使得它适用于任意多边形计算单元.为了得到定常解,采用一种显式的四步Runge-Kutta迭代方法对时间进行积分.根据流场参数的变化梯度确定加密边,由加密准则进行自适应网格剖分,然后得到分布合理的加密过后的网格.求解二维Euler方程,对NACA0012翼型进行了数值模拟,通过对自适应前后的数值解的对比,说明所建立的方法是正确的.