稀疏矩阵

然后对稀疏矩阵变换器的换流、调制策略进行了详细分析；最后，利用Matlab软件对稀疏矩阵变换器进行建模仿真。

Finally , set up the simulation model of the sparse matrix converter by means of Matlab .

基于CIM的稀疏矩阵模型设计

Design of a CIM-BASED sparse matrix model

处理稀疏矩阵的C语言链式三元组表法

Chaining Three-Element Technique on C Language to Process Sparse Matrices

利用EXCEL快速生成MATLAB大规模稀疏矩阵

To Create Matlab Large-scale Sparse Matrices Rapidly by Using Excel

大型PMS对应的Markov模型的转移概率矩阵通常为大型稀疏矩阵。

Normally , transition probability matrix in Markov model of the large PMS is large and sparse .

ADAMS采用拉格朗日动力学方程，辅以刚性积分算法以及稀疏矩阵技术来求解模型。

ADAMS solves the model by adopting Lagrange dynamics equation and complementing with rigidity integ-ral algorithm and sparse matrix technology .

本文主要设计了基于FPGA实现稀疏矩阵分解的硬件结构。本文首先介绍了稀疏矩阵的特点和研究稀疏矩阵分解的意义，接着讨论了稀疏矩阵各种快速算法并给出了本文所采用的方法。

This dissertation mainly discusses the realization of sparse matrix factorization based on FPGA . Firstly , the characteristics and research value of sparse matrix are introduced .

在仿真计算中，采用隐式Euler算法、线性迭代方法和稀疏矩阵算法求解由物理模型所得的非线性微分方程组。

The simulation method uses Euler 's method for the transient terms and solves the equations using relaxation and a diagonal matrix algorithm .

本文分析了大型稀疏矩阵的结构特点，并利用CSR压缩存储方式对稀疏矩阵进行压缩存储，从而节省了稀疏矩阵的存储空间。

By analyzing the large sparse matrix and storing it with CSR , the storage space of large sparse matrix was saved .

采用ADI与高阶紧致差分相结合的方法计算大型非对称稀疏矩阵，并实现了该算法在半导体器件模拟中的应用。

The solution using a spline procedure , SADI , and a high-order compact finite difference ( HOC ) method is presented for the hydrodynamic ( HD ) model for semiconductor device simulation .

用Lanczos方法解高阶稀疏矩阵广义特征值问题的某些经验

Some experience in using lanczos ' method to solve the high order sparse matrix generalized eigenvalue problem

本文通过利用稀疏矩阵及递归的方法，提出了一种PERT图在计算机中的链式存贮的存贮方法和计算方法的改进方案。

This paper mainly introduces an improvement of linked representation of PERT graph and its method of calculation in the comuter .

建立了一套稀疏矩阵的高密压缩技术，并结合这一技术对SOR迭代法进行了改进，大大提高了程序的运行效率。

A technology for compressing the sparse matrix consumedly was created . The SOR iteration was improved through combing with this technology .

本文提出在LU分解中按填入数最少原则（或按长运算次数最少原则），对结构不对称稀疏矩阵[A]进行重排序的一种方法&有向图消去法。

The paper presents an elimination approach to the oriented graph which is used to reorder the structure of the asymmetric sparse-matrix according to the rule of minimal fills or minimal long operations in LU factorization .

选择共轭梯度法解决由有限元法形成的大型稀疏矩阵方程，应用FORTRAN语言编制了数值模拟系统软件。

Conjugate gradient method were choosed to solve the large sparse matrix equations induced by the FEM and computer language FORTRAN was used to programme the numerical simulation system software .

该算法的一个主要特点是算法产生的矩阵序列保持目标函数的Hessian阵是一个具有类似于块对角状的稀疏矩阵。

Thereby , the Hessian matrix of the objective function being similar to a block diagonal sparse matrix is maintained by the updated matrix sequence .

详细介绍了Excel与Matlab的连接方法，并实现了Excel向Matlab传送大规模稀疏矩阵，充分利用了两个软件的优点，方法简便、独特。

The article describes the connecting way of Excel and Matlab and realizes the transferring of large-scale sparse matrices from Excel to Matlab , it makes fully use of the advantage of two software with a simple and particular method .

利用DIA存储格式把带状稀疏矩阵转换成向量，同时在迭代开始前把全部数据传递给GPU，迭代过程中尽量避免与CPU的数据通信。

By using the DIA storage format to convert the band sparse matrix to vectors , we delivered all the data to GPU before start of the iteration and only communicated a little data with CPU .

最优嵌入问题是从稀疏矩阵计算，纠错码，数据结构，VLSI及分子生物学等中提取出来的数学模型，有着广泛的应用背景。

Optimal embedding ( labelling ) problem for graphs arises from the circuit lay-out of VLSI designs , interconnection networks , sparse matrix computations , error-correcting code designs , data structures , biology , etc , which has extensive back-grounds .

提出了利用小波变换（WT）、非负稀疏矩阵分解（NMFs）和Fisher线性判别（FLD）来进行人脸识别。

This paper combines Wavelet Transformation ( WT ), Non-negative Matrix Factorization with sparseness constraints ( NMFs ), and Fisher 's Linear Discriminant ( FLD ) to extract features for face recognition .

Gallager在1962年提出的低密度奇偶校验码（Low-DensityParity-CheckCodes，LDPC）是一类可以用稀疏矩阵或二分图定义的线性分组码。

Low-density parity-check codes ( LDPC ) first discovered by Gallager in 1962 are a class of linear block error-correcting codes that can be defined by the very sparse parity-check matrix or the bipartite graph .

超松弛迭代法（简称SOR法）是解决大型稀疏矩阵方程组的有效方法之一，是一种一阶线性定常迭代法。

The successive over-relaxation method ( SOR method ) is one of the effective methods to solve large sparse matrix eguation sets . It is a kind of one-order linear constant iterative method .

接着研究了两种求解大型稀疏矩阵的解法&共轭梯度法和LU分解法，也从计算精度、计算内存需求及计算效率分析了这两种方法的特点。

Then , I perform two methods to solve large scale linear sparse systems for seismic modeling & Conjugate gradient method and LU method . I also compare and analyze these two methods on several aspects , such as computational accuracy , memory requirement and CPU run time .

该模式以BOD5总量为控制对象，任一河流断面的BOD5输入量在此响应矩阵中仅出现一次，从而使响应关系由下三角矩阵改变为每一行均含三个变量的稀疏矩阵。

Based upon total load control , a BOD5 input to any river reach appears in the response matrix only once which changes the response from a lower triangular matrix to a sparse matrix in which any single row has three variables .

本文以Aki和Richards近似公式为基础，将最小平方问题转化为大型稀疏矩阵的求解问题，直接反演出纵、横波速度；

Based on approximate formula presented by Mr. Aki and Richards , the paper transforms the least square issues into finding the solution of large-scale sparse matrix , directly makes the inversion of P wave and S wave velocities ;

其中，基波计算部分采用PQ分解法，并采用稀疏矩阵和对称矩阵处理技术，利用三相因子表求解，在很大程度上节省了内存，减少了计算量；

In this algorithm , the three-phase fundamental power flow is solved by the PQ decoupled method in which the sparse and symmetrical matrix handling technology and three-phase factor table are used , thus memory is saved and computing quantity is decreased .

选择适合求解大型不对称稀疏矩阵的MILU-ORTHOMIN（预处理正交极小化）方法，是预处理共轭梯度法的一种，是MILU（修正不完全因子分解）法和ORTHOMIN（正交极小化）法的有机组合；

The modified incomplete factorization ( MILU ) - orthogonal minimization ( ORTHOMIN ) method can be used to calculate large asymmetric sparse matrix as one of the pretreatment conjugate gradient methods .

继而针对稀疏矩阵所代表的点滴流形状的流体，应用C++语言编写的Maya插件，对两相流流体进行实时仿真，最后对所建立的整个场景进行渲染并输出动画。

Four , the Maya plug-in unit composed by C + + language is applied to stimulate the two phase flow in real-time in accordance with the fluid represented by the sparse matrix . Finally , the whole scene is embellished and the animation is output .

同时，将广义Hardmard矩阵乘积的性质，应用于解某种特定的大型稀疏矩阵的线性方程组问题。

To show the usefulness of applying these properties of the generalized Hadamard products , the solution of large-scale linear system problems with sparse matrix is considered .

采用改进的ICCG（不完全乔尔斯基共轭梯度）方法，对一维变带宽存储的大型复稀疏矩阵进行了求解，得到了随钻电磁波测井响应。

The large scale complex equations are stored by one-dimensional variant-banded and are solved by the method of improved incomplete Choleskey conjugate gradient and the response of electromagnetic LWD tool is deserved .