数值算法

To speed up the convergence rate of the iterative algor...

用所提出的耦合算法分析算例，得到的结果与有限元法、边界元法和另一种耦合算法的数值计算结果一致，验证了这种算法的正确性和可行性。

Numerical experiments show that the new algorithm is not only an all-purpose algorithm, but also a high-performance algorithm with better results than others.

数值实验表明，新算法不仅是一种通用的算法，而且与已有算法的计算结果相比，其解的精确度也最好。

People have proposed many different control method to realize the chaos synchronization. This article has conducted the thorough research to the chaotic neural network, introduced the chaos neural network synchronization method. Based on the linear matrix inequality technology and Lyapunov stability theory, we have Studied the synchronized controller design method of the global synchronization, the non-linear global synchronization and the global exponential synchronization. Here we have proposed one kind of synchronization algorithm, through which might obtain the gain matrix quickly and automatize.

本文对混沌神经网络进行了深入的研究，介绍了混沌神经网络同步方法的发展及研究现状，基于线性矩阵不等式技术和李雅普诺夫稳定性理论，分别研究了全局同步、非线性全局同步和全局指数同步的同步控制器实现方法，提出了一种实现混沌神经网络同步的算法，通过该算法可以快速得到增益矩阵的形式，实现混沌系统的同步控制，数值仿真验证了算法的有效性。

In this paper, an extended divide and conquer algorithm is intended proposed, which is for solving the real symmetric band generalized eigenvalue problem under distributing environment Eigenvalue partition theorem is presented and proved Based on divide and conquer by extension, this algorithm computes generalized eigenpaires of symmetric band matrix pencil by bisection and generalized Rayleigh quotient iteration Theoretic analysis and numerical results show that this algorithm is better than the classic software package LAPACK when bandwidth is small and the scale is large Combined with multisection, which has good parallelism, it got good effects under distributed environments

提出了分布式环境下计算对称带状广义特征值问题的一种扩展分治算法，给出了特征值分割定理及其证明算法在扩展分治的基础上，利用二分压缩结合广义Rayleigh商迭代计算广义特征对理论分析和数值实验表明，对于窄带宽大规模的广义特征值问题，该分治算法明显优于LAPACK软件包结合并行性好的多分法，在分布式环境下获得了很好的并行效果1 引言本文研究了对称带状广义特征值问题Ax =λBx ( 1)的并行计算，其中，A ，B均为半带宽为r的n阶实对称带状矩阵且其中之一是正定的本文总假设B是正定的求解此问题有两种传统方法，第1种方法是通过计算矩阵B的Cholesky分解，将问题( 1)转化为标准特征值问题[1～3] ，进一步

The concept of genetic algorithm, its configuration, operators and existing problems are introduced in this paper. A new algorithm for clustering analysis is presented based on genetic algorithm. There are two characteristics in our approaches. Firstly, the algorithm is the general-purposed and our clustering analyzer can cluster large data set with mixed numeric and categorical attributes.

4介绍遗传算法的概念、算法结构、运算算子及主要存在问题；提出了一种基于遗传算法的聚类新方法MGA，该方法有两个主要优点：一是通用性强，可以对包含数值属性和符号属性的大数据集进行聚类；二是提高了数据挖掘的效率与质量。

Apply the thought of differential and integral calculus to the problem of global optimization, imitate the algorithm of digital approximation of definite integral and construct a category of algorithms that can solve global optimization with great brevity and utility.

把微积分的基本思想应用到最优化问题中，仿照定积分的数值逼近算法构造出一类简洁实用的全局优化算法-积分选择算法。

The 2-D transfer function of the bistatic SAR is obtained. Azimuth position distortion along range existing in the original alogrithm is corrected by mapping from scene coordinate to azimuth-slant distance coordinate. Bistatic SAR date can still be processed by a mono-static SAR processor. The efficaciousness of the proposed algorithm is verified through point target simulations.

算法通过数值计算得到双基地二维传递函数，通过场景坐标系向方位－斜距平面的映射校正了原算法中存在的沿距离变化的方位位置畸变，进而来用常规单基地SAR成像处理流程实现对双基地SAR回波数据的成像处理，最后通过点目标仿真验证了算法的有效性。

A new method to solve the combinatorial optimization problems with mixed evolutionary algorithm based on natural scale enumerative system coding was proposed in this paper. Elitism strategy, one-point crossover and Gaussian mutation were used in the algorithm. An ingenious method for operation among big parameters was designed, which can avoid overflow and simplify the operation.

在自然进位制编码的基础上，算法采用了遗传算法的单点交叉算子和进化规划的高斯扰动算子，并运用了精英保留策略；算法实现时采用逐位运算法实现大数值运算，避免了运算溢出，减少了运算量。

From the ergodicity and the regularity of chaotic behavior,it is conclude that the algorithm has the asymptotical convergence property.

基于变分不等式的等价形式，构造了一种混沌搜索算法来直接求解变分不等式问题，根据混沌行为的遍历性和规则性，可以得到算法的渐进收敛性。数值试验结果表明，该算法是一个有效的方法

The following are the results: 1. An abstract meteorological phase parallel model for numerical weather prediction is developed as follows:〓 In this equation, parallel wall clock time related with computing grid numbers, load imbalance, communication and synchronization overhead is given out. It can be used to design and analysis the parallel algorithm.

该研究工作的主要结果如下： 1 发展提出了一个适用于气象数值预报模式的阶段并行模型MPPM：〓它表明了并行计算时间与数值模式的计算格点量、负载平衡因子、通信开销、同步开销及并行开销等之间的关系，为设计和分析数值预报模式的并行算法提供了理论依据。

The split between the two groups can hardly be papered over.

这两个团体间的分歧难以掩饰。

This approach not only encourages a greater number of responses, but minimizes the likelihood of stale groupthink.

这种做法不仅鼓励了更多的反应，而且减少跟风的可能性。