An efficient algorithm for hypersonic viscous flows

The CSCM-S algorithm proposed by Lombard et al. is a very attractive tool for solving multidimensional Euler and Navier-Stokes equations. However, it is not economical due to the use of global sweeps in the whole computational domain. In this paper we suggest a modified strategy, which combines a single-marching technique for supersonic dominated region with a multi-sweep procedure for pure subsonic and complex flowfield. The new algorithm may save significantly CPU time and is more suitable for engineering applications. The project supported by National Natural Science Foundation of China     

计算具有区间参数结构的固有频率的优化方法

基于区间函数的单向包含性质，把具有区间非确定参数结构的固有频率所在区间范围问题 转化成两个全局优化问题，并采用一种实数编码遗传算法求取问题的全局解. 用一种能够求 得剪切型结构和弹簧质量系统特征值范围精确解的单调分析方法进行检验. 在一 些文献中，直接采用区间数运算法则和有限元法得到结构区间刚度阵和区间质量阵，并把关 于该区间刚度阵和区间质量阵的广义区间特征值问题的特征值区间作为待求的非确定性结构 的特征值所在的区间范围，该方法易于扩大问题的解域. 算例表明，可望得到结构 固有频率区间范围的准确解.     

条形平板中裂纹识别的遗传算法

研究条形平板中平行于上下边界的单一裂纹的识别问题。给出了当一声波入射这一条形平板时，裂纹参数与反射、透射系数之间的关系式，并应用遗传算法对裂纹进行了定量识别。计算结果表明本文给出的方法具有较好的识别精度。     

部分输入未知条件下结构动力复合反演的分解算法

在充分利用部分输入已确知而部分输入未知的激励特性的基础上，提出了结构动力复合反演的分解算法，该算法从源头上消除了迭代过程中参数识别与荷载反演的相互影响，降低了问题的计算规模。对于线性参数系统，该算法不经过任何迭代计算即可一次性完成结构参数识别及荷载反演。将其与松弛法结合，可解决非线性参数系统的识别问题，与文献[4]的方法比较，其收敛速度有显著提高。     

On the solution of plane boundary-value problems of elasticity in a rectangle

An algorithm for solving plane boundary-value problems of elasticity for a rectangular domain is expounded. The algorithm is based on a complex-valued representation of the general solution to the differential equations of the plane problem and on the use of Lagrange polynomials to satisfy the boundary conditions. The algorithm can quite easily be implemented in a computer program. This is probably the simplest way of solving boundary-value problems of this class __________ Translated from Prikladnaya Mekhanika, Vol. 42, No. 1, pp. 97–102, January 2006.     

On numerical solution of dynamic problems in the theory of reinforced shells

Dynamic problems for cylindrical shells reinforced with discrete ribs are examined. A numerical algorithm based on Richardson extrapolation is developed. Specific problems are solved, and the results are analyzed __________ Translated from Prikladnaya Mekhanika, Vol. 42, No. 5, pp. 50–56, May 2006.     

一种用于分散化滤波的融合算法及其在组合导航系统中的应用

提出了一种局部估计相关时的最优融合算法，其特点是加权矩阵是对角阵，即全局估计中的各分量分别是局部估计中相应分量的线性组合，在此融合算法中局部估计是最优的，并且不存在由主滤波器向子滤波器的信息反馈，因此具有较好的客措性。给出了最优加权矩阵的存在性和唯一性的证明，并通过与现有类似融合算法的对比，表明该算法具有计算量小的优点。最后，以组合导航系统为例作了仿真计算，结果表明该融合算法是可行的。     

基于非结构化同位网格的SIMPLE算法

通过基于非结构化网格的有限体积法对二维稳态Navier—Stokes方程进行了数值求解。其中对流项采用延迟修正的二阶格式进行离散；扩散项的离散采用二阶中心差分格式；对于压力-速度耦合利用SIMPLE算法进行处理；计算节点的布置采用同位网格技术，界面流速通过动量插值确定。本文对方腔驱动流、倾斜腔驱动流和圆柱外部绕流问题进行了计算，讨论了非结构化同位网格有限体积法在实现SIMPLE算法时，迭代次数与欠松弛系数的关系、不同网格情况的收敛性、同结构化网格的对比以及流场尾迹结构。通过和以往结果比较可知，本文的方法是准确和可信的。     

结构拓扑优化设计的三角网格进化法

针对进化或拓扑优化方法的不足，提出了一种基于遗传算法的新型进化式拓扑优化方法－－三角网格进化法，该方法不仅能够同时进行拓扑，形状与截面变量优化设计，而且在优化过程中实现了退化和进化的统一，提高了优化效率。另外本文还首次对结构类型变量进行了优化计算，取得了有益的结果。最后几个数值算例证明了本方法的可行性和有效性。     

Fast parallel algorithm for three-dimensional distance-driven model in iterative computed tomography reconstruction

The projection matrix model is used to describe the physical relationship between reconstructed object and projection.Such a model has a strong influence on projection and backprojection,two vital operations in iterative computed tomographic reconstruction.The distance-driven model(DDM) is a state-of-the-art technology that simulates forward and back projections.This model has a low computational complexity and a relatively high spatial resolution;however,it includes only a few methods in a parallel operation with a matched model scheme.This study introduces a fast and parallelizable algorithm to improve the traditional DDM for computing the parallel projection and backprojection operations.Our proposed model has been implemented on a GPU(graphic processing unit) platform and has achieved satisfactory computational efficiency with no approximation.The runtime for the projection and backprojection operations with our model is approximately 4.5 s and 10.5 s per loop,respectively,with an image size of 256×256×256 and 360 projections with a size of 512×512.We compare several general algorithms that have been proposed for maximizing GPU efficiency by using the unmatched projection/backprojection models in a parallel computation.The imaging resolution is not sacrificed and remains accurate during computed tomographic reconstruction.