中国科学院数学与系统科学研究院期刊网

数值计算与计算机应用 1999年 20卷

Please wait a minute...
选择: 显示/隐藏图片
1. 欧拉方程的velicity表示和辛积分(Ⅱ)
陈旻
数值计算与计算机应用    1999, 20 (1): 1-9.   DOI: 10.12288/szjs.1999.1.1
摘要755)      PDF(pc) (278KB)(291)    收藏
In this paper the author applies the implicit midpoint scheme (IM) to her Hamiltonian system for large time, with the fixed time step (t) to be 0.1. Two velicitieswhose initial data diner by a gradient yield the same velocity field for all time: the author has verified this in a numerical experiment using the IM scheme, up to t = 2.
参考文献 | 相关文章 | 多维度评价
2. 三维有限元计算中任意剖面上的数据处理
田春松,胡健伟
数值计算与计算机应用    1999, 20 (1): 10-20.   DOI: 10.12288/szjs.1999.1.10
摘要708)      PDF(pc) (539KB)(261)    收藏
The postprocessing for 3-D finite element methods is discussed in detail. For anarbitrary section in 3-D domain, the data processing of planar mesh is presented inorder to draw the contour map, the shaded map, the surface map easily and exactly.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(1)
3. 边界元计算中一种新的积分方法
侯劲松,王泽毅
数值计算与计算机应用    1999, 20 (1): 21-27.   DOI: 10.12288/szjs.1999.1.21
摘要780)      PDF(pc) (281KB)(259)    收藏
In three dimensional boundary element calculation, the boundary integration isthe most take-consuming work. In this paper, a new approach called parameterizedintegration method is presented. It calculates values of a multivariate function insteadof traditional Gauss integral and saves integral time.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(10) CSCD(1)
4. 钝舵绕流不同湍流模型数值模拟比较研究
马汉东,李素循,吴礼义
数值计算与计算机应用    1999, 20 (1): 28-34.   DOI: 10.12288/szjs.1999.1.28
摘要747)      PDF(pc) (292KB)(294)    收藏
The Reynolds-averaged Navier-Stokes equations are solved numerically for supersonic flow over a blunt fin mounted on a flat plate with Roe third-order flux differencesplitting scheme. Turbulence eddy-viscousity is computed by Baldwin-Lomax model,Johnson-King model and Launder-Sharma k - model. The pressure distributions arein good agreement with experimental results. Some features, such as two reversed-flowsupersonic zones and secondary vortex structure are predicted well. The comparativestudy suggestes: (1) The Wilcox compressibility correction is more potent than sarkarcorrection for the case. (2) Johnson-King model provides a better overall simulation.(3) The performances of three different kinds of turbulence models are essentially thesame, but the computation efforts increase one by one.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(2)
5. 壳体上的C~1插值
徐国良
数值计算与计算机应用    1999, 20 (1): 35-45.   DOI: 10.12288/szjs.1999.1.35
摘要765)      PDF(pc) (380KB)(261)    收藏
We propose an approach to construct a C1 function on a fat surface domain, asurface that has thickness, by piecewise rational functions defined on a collection ofirregular triangular prisms. We show that the interpolation function has algebraic precision 2. Examples that show the effectiveness of the scheme is presented.
参考文献 | 相关文章 | 多维度评价
6. H~1(R)H~1(R)空间中广义二维可分离微商算子样条多分辨分析
朱孟芝,邓中兴,邓彩霞
数值计算与计算机应用    1999, 20 (1): 46-56.   DOI: 10.12288/szjs.1999.1.46
摘要818)      PDF(pc) (294KB)(238)    收藏
This paper established a multiresolution analysis of two-dimension seperable differential operator spline with non-polynomial type in the reproducing space H1 (R)H1 (R) )realized the expansion of function according to the system of Dyadic extension andtranslation, and its coefficients of expansion was demonstrated by the value of functionaccurately. Moreover, this paper provided reasonable approach and efficient methodfor extending the theory of two-dimension operator spline with non-polynomial type.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(12)
7. 有限元大型二次规划解的一种新算法
孙秦,范建华,徐晓飞
数值计算与计算机应用    1999, 20 (1): 57-61.   DOI: 10.12288/szjs.1999.1.57
摘要760)      PDF(pc) (218KB)(285)    收藏
In order to efficiently solve large-scale linear and complementary equation systemdeduced by quadratic programming solution of elastoplastic finite element analysis, thispresent paper works out a new algorithm which can obviate insignificant data storeand computation, and so make the solution procedure much more efficient and practical for microcomputers. The key point of this proposed algorithm lies in elastoplasticnature only existing some local area in a studied domain. So characteristic and noncharacteristic areas should be estimated in advance, then routine computational nodesand net is designed and numbered according to the area division. This algorithm alsopresents consecutive detailed procedure for storage and run-time saving strategies oflarge-scale sparse systematic matrices, which are linear store and triangular decomposition techniques. This algorithm is checked by two test examples for its high efficiency, good numerical stability and advisability.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(6)
8. 非整数阶微积分的滤波特性及数值算法
陈遵德,陈富贵
数值计算与计算机应用    1999, 20 (1): 62-69.   DOI: 10.12288/szjs.1999.1.62
摘要764)      PDF(pc) (304KB)(442)    收藏
This paper, with the use of Fourier transform and the filtering theory of linear timeinvariant system, reaches a conclusion that there are high-pass filter characteristicsin derivative and low-pass filter characteristics in integration after analysising fromthe view point of filtering. On the basis above, the author explores the numericalalgorithm of calculus-FFT algorithms and its adaptable condition by combining thesampling theorem of time with frequency domain, which is popularized to the numericalcomputation (NC) of non-integral-order calculus. The analysis and results of NC showthat the algorithm is convenient, fast, and applicable for the NC of low-order calculus,and specially suited to the engineering computation. It will aid in the NC, practicalutilization and error analysis of non-integer-order calculus.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(47) CSCD(3)
9. 多步Runge-Kutta方法的PDIRK实现
廖文远,李庆扬
数值计算与计算机应用    1999, 20 (1): 70-80.   DOI: 10.12288/szjs.1999.1.70
摘要737)      PDF(pc) (417KB)(329)    收藏
In this paper, we presented some parallel diagonal implicit Runge-Kutta (PDIRK)iterative algorithms for a class of multistep Runge-Kutta methods presented by Butcherin 1960's. Also in this paper we reached and proved some conclusions on convergence,error estimation and stability of these algorithms. Numerical experiments showed thatthese algorithms are stable, convergent and error can be controlled if m (iteration times)is large enough.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(1)
10. 一种预测火灾事故的数值模拟方法
魏伟
数值计算与计算机应用    1999, 20 (2): 81-87.   DOI: 10.12288/szjs.1999.2.81
摘要751)      PDF(pc) (244KB)(263)    收藏
It has a lot of advantages to forecast three fire accident indexs, that is, numberof accidents, number of the wounded and economical loss by the use of grey systemtheory corning out in the 1980s, which is beyond compare to other forecast methods.This paper first describes the method for forecasting fire accidents by the use of greysystem theory' an example is given to use a microcomputer aided forecasting, whichshows application of the method.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(6)
11. 指数拟合的Runge-Kutta公式
阮保庚
数值计算与计算机应用    1999, 20 (2): 88-97.   DOI: 10.12288/szjs.1999.2.88
摘要808)      PDF(pc) (293KB)(159)    收藏
A class of RK methods consistent of order at least 2 is offered so that the unity of thetwo-stage Gauss, Radau and Lobatte formulae is achieved. Based on the class of axmethods exponentially fitted ax formulae of order 2 with two parameters and of order3 with one parameter are given, which are not only A-stable but also algebraicallystable, and therefore suitable for integration with a large stepsize for fast decayingcomponents of linear and nonlinear stiff systems.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(12) CSCD(3)
12. 一类求解刚性常微分方程的多步插值法及其平行实现
廖文远,李庆扬
数值计算与计算机应用    1999, 20 (2): 98-106.   DOI: 10.12288/szjs.1999.2.98
摘要942)      PDF(pc) (336KB)(278)    收藏
In this paper, we combine implicit RK method with implicit linear multistep methodto propose a class of multistep interpolation methods which make use of the derivativesat the point of t.--k+j (j ~ 1, 2,''', k -- 1) and baize on general multistep ax methodproposed in [1]. By choosing free parameters in these methods we can make themstable, parallable. The other advantages are their high orders and less computation.Numerical experiments show that these methods are more efficient in solving stiff ODEsthan implicit ax method, implicit linear multistep method and multistep RK method
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(5)
13. 求解块角形线性规划问题的保形算法
赵实
数值计算与计算机应用    1999, 20 (2): 107-123.   DOI: 10.12288/szjs.1999.2.107
摘要800)      PDF(pc) (458KB)(240)    收藏
The algorithm is introduced in this paper, it not only applies simplex method butalso remains the original form of the constraint matrix of the problem in the processof solving the block-angular structure. It has the following advantages:1. The iteration times of it are the same as that of simplex method.2. The calculation of iteration is lesser.3. In every time of iteration, the calculation scale remains the same.4. It can proofread the error of calculation or result to obtain the solution exactly.
参考文献 | 相关文章 | 多维度评价
14. 矩形区域上仿射变分不等式的一个直接法
李董辉,杨余飞
数值计算与计算机应用    1999, 20 (2): 124-130.   DOI: 10.12288/szjs.1999.2.124
摘要782)      PDF(pc) (252KB)(227)    收藏
In this paper we present a simple and practical direct method solving a class ofaffine variational inequalities on rectangle domain. The algorithm is of polynomialtime. And its multiplication work is no more than that of Gauss elimination withpivoting for solving linear system of equations when the variational inequality is alinear complementarity problem..
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(12)
15. 一类本身具有预条件功能的ABS型共轭梯度算法
李春光,游兆永
数值计算与计算机应用    1999, 20 (2): 131-137.   DOI: 10.12288/szjs.1999.2.131
摘要729)      PDF(pc) (211KB)(300)    收藏
This paper presents a new CG-type algorithm for solving large linear systems. It isobtained from a subclass of the ABS algorithm-Voyevodin's CG method by choosingthe parameter matrix B in some special ways. Having preconditioning properties, thematrix B makes the new algorithm converge fast. The convergence analysis is given.Several ways for choosing B, which are similar to the polynomial preconditioing, arediscussed. Numerical tests indicate that the new algorithm is effective and competitive.Besides, it is suitable for parallel architectectures.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(1) CSCD(1)
16. 一种二维数值网格的构造方法
徐涛,水鸿寿
数值计算与计算机应用    1999, 20 (2): 138-152.   DOI: 10.12288/szjs.1999.2.138
摘要759)      PDF(pc) (663KB)(240)    收藏
The algebraic expressions describing the geometric properties of mesh curves aregiven. The control functions concering above properties are introduced and functionalsshowing those properties are constructed. Based on a theorem, we get the concretecorresponding relation of the control functions with the mesh properties when thefuctional taken minimum. By the numerical solution of Euler equation obtained fromfunctional minimum, we get the needed grids. This method of grids generation has aquit derict geometric meaning, and its inner control is flexible. Particularly not onlythe adaptability with physical preoblem is concerned, but also the adaptability with thereal physical region. The numerical experiments verified the efficiency of our method.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(5) CSCD(1)
17. 求解变量带简单界约束的非线性规划问题的障碍函数法
马昌凤
数值计算与计算机应用    1999, 20 (2): 153-160.   DOI: 10.12288/szjs.1999.2.153
摘要735)      PDF(pc) (239KB)(297)    收藏
In this paper, we present a barrier function algorithm for bound constrainted minimization. Under milder conditions, we prove the convergence of the algorithm. Inaddition, we also carry out some numerical experiments and obtain fairly good numerical results.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(3) CSCD(1)
18. 非对称广义特征值问题的并行同伦-行列式算法
薛长峰,周树荃
数值计算与计算机应用    1999, 20 (3): 161-166.   DOI: 10.12288/szjs.1999.3.161
摘要769)      收藏
Parallel computation for generalized eigenvalue problem is one of the fundalmentalproblems in large-scale engineering computation. Zhou Shuquan, Deng Shaozhong andZeng Lan have made a systematic and deepgoing study of parallel computation forsymmetric generalized eigenvalue problem. So far there is very little study of parallelcomputation for nonsymmetric case. In this paper we apply homotopy continuationalgorithm to parallel computation for nonsymmetric generalized eigenvalue problemand present parallel homotopy-determinant algorithm. Numerical results show that ithas advantage of high parallel efficiency and fast covergence speed.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(8)
19. 非线性奇异摄动问题的高精度算法
吴雄华,谭志海
数值计算与计算机应用    1999, 20 (3): 167-173.   DOI: 10.12288/szjs.1999.3.167
摘要746)      收藏
In this paper, the method with high-order accuracy was presented for a kind ofnonlinear sigular perturbation problems. The numerical tests demonstrate that only afew grid points are enough for the numerical results with high-order accuracy in ourmethod.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(6) CSCD(1)
20. 自适应模糊滤波方法
谢承迪
数值计算与计算机应用    1999, 20 (3): 174-183.   DOI: 10.12288/szjs.1999.3.174
摘要700)      PDF(pc) (316KB)(275)    收藏
The fuzzy estimation method is developed based on theory of probability and statistics and the results are utilizied in state estimation of stochastic system. By means ofthe approach, a fuzzy filtering method is formed to improve the filtering effects.
参考文献 | 相关文章 | 多维度评价
21. 关于采用流水线方式进行一簇递推关系式的并行计算
张林波
数值计算与计算机应用    1999, 20 (3): 184-191.   DOI: 10.12288/szjs.1999.3.184
摘要718)      PDF(pc) (341KB)(299)    收藏
In this paper parallel implementation issues concerning computation of a set ofrecurrences on distributed memory parallel systems are discussed. It is a commonproblem when solving partial differential equations on distributed memory systems,especially in computational fluid dynamics. For example, solution of a set of tri-diagonalsystems of equations or Guauss-Seidel relaxations on finite difference systems all leadto this kind of computation. We emphasize on the idea of doing the computations in apipelined fashion and we show through analysis and numerical examples that by usingproperly chosen parameters, pipelined implementation often yields much better parallelefficiency with respect to other commonly used methods.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(28) CSCD(4)
22. 一种快速生成2~k3型Hadamard矩阵的算法
陈勤
数值计算与计算机应用    1999, 20 (3): 192-198.   DOI: 10.12288/szjs.1999.3.192
摘要736)      PDF(pc) (229KB)(209)    收藏
Hadamard matrix is widely applied in digital image processing, encoding theory,digital communication, multiple bits diffusion code, block design and etc. Production ofHadamard matrix with 2k has been solved, but production of other Hadamard mains isvery difficult. In this paper, the best deviation matrix is proposed, and some propertiesand a relation with Hadamard matrix are obtained. According to these results, afast production algorithm of Hadamard matrix with 2k3 is given, and its algorithmcomplexity is analysed.
参考文献 | 相关文章 | 多维度评价
23. 求解混合0-1线性规划问题的一个分支定界算法
倪明放,江泽焱
数值计算与计算机应用    1999, 20 (3): 199-204.   DOI: 10.12288/szjs.1999.3.199
摘要766)      PDF(pc) (180KB)(477)    收藏
In this paper, a branch and bound algorithm for solving mixed 0-1 linear programming problem is presented and the effectiveness of the algorithm is illustrated bycomputation examples.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(26) CSCD(3)
24. 任意散乱数据光滑超曲面拟合的二步法
唐月红,李卫国,郭红英
数值计算与计算机应用    1999, 20 (3): 205-214.   DOI: 10.12288/szjs.1999.3.205
摘要723)      PDF(pc) (404KB)(248)    收藏
In this paper, a fitting method for arbitrary multivariate scattered data is presented.The method consists of uniforming the set of arbitrary 4dimensional scattered data byusing local weighted least square, evaluating the values of partial derivatives and minedpartial derivatives over the cube-gridded-points only by using those gridded-point data,followed by interpolating smoothly and quickly those gridded-point data. Good resultshave been gotten by computing several known functions.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(23) CSCD(1)
25. 关于可正定化矩阵的判定
王伟贤,王志伟
数值计算与计算机应用    1999, 20 (3): 215-222.   DOI: 10.12288/szjs.1999.3.215
摘要738)      PDF(pc) (244KB)(322)    收藏
In this paper, we study the method of ascertaining positive-definable matrix. Weimprove the results in a paper by Hu Jiagan & Liu Xingping, and give a new method.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(2)
26. 基于Samelson逆的矩阵ε-算法和η算法及对矩阵指数的应用
顾传青
数值计算与计算机应用    1999, 20 (3): 223-230.   DOI: 10.12288/szjs.1999.3.223
摘要826)      PDF(pc) (260KB)(394)    收藏
The Samelson inverse mains Pade approximation (SMPA) is introduced in [3]. Bymeans of Samelson invrese for matrices, s-algorithm and n-algorithm, which are used asthe computation for SMPA, are established. The well-known Wynn identity for SMPAis given on the basis of ε-algorithm. As a important application, matrix ε-algorithm formains exponeatial functions is established. Stability analysis for three approximatingformulas, which is often used, point out that matrix ε-new algorithm is stable, on thecondition of characteristic value of the matrix discussed is less than or equal to zero.
参考文献 | 相关文章 | 多维度评价
被引次数: CSCD(2)
27. Laplace变换的数值反演方法的研究
单华宁
数值计算与计算机应用    1999, 20 (3): 231-236.   DOI: 10.12288/szjs.1999.3.231
摘要800)      PDF(pc) (171KB)(180)    收藏
Based on discussing the method of accelerating the convergence of infinite sequences,the study on the method for numerical inversion of Laplace transforms has been made.A new method of numerical inversion of Laplace transforms is presented. This methodnot only improves the accuracy, but also has a better numerical stability.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(19) CSCD(1)
28. 二维三温热传导方程组的九点差分格式
符尚武,付汉清,沈隆钧
数值计算与计算机应用    1999, 20 (3): 237-240.   DOI: 10.12288/szjs.1999.3.237
摘要1293)      PDF(pc) (150KB)(559)    收藏
In this paper, we present a nine-point difference scheme for two-dimensional equa- tions of heat conduction with three-temperature. Finally, the calculated result arecompared with the five-point scheme.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(30) CSCD(8)
29. 检验0,1数字序列随机性的多重异或法
罗平
数值计算与计算机应用    1999, 20 (4): 241-248.   DOI: 10.12288/szjs.1999.4.241
摘要838)      PDF(pc) (261KB)(328)    收藏
This paper is concerned with testing 0, 1 digital sequence for its randomness. Itsprobability model is set up and probability distribution law is deduced. Based on them,the multiple XOR method for testing randomness of a digital sequence is put forward.Digital simulation done on computer shows that this method is effective.
参考文献 | 相关文章 | 多维度评价
30. 关于时滞微分方程初值问题的一类并行算法
甘四清,张诚坚
数值计算与计算机应用    1999, 20 (4): 249-254.   DOI: 10.12288/szjs.1999.4.249
摘要805)      PDF(pc) (178KB)(441)    收藏
Based on the algorithms which was presented by Li Shou-fo and Gan Si-qing[1],a class of parallel algorithms for initial value problems in DDEs is constructed. Thenumerical results show the effectiveness of these parallel algorithms for linear and nonlinear DDEs.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(12) CSCD(3)
31. Krylov子空间方法在一类PDE上的数值实验与分析
曹建文
数值计算与计算机应用    1999, 20 (4): 255-265.   DOI: 10.12288/szjs.1999.4.255
摘要791)      PDF(pc) (483KB)(379)    收藏
In this paper the analyses of various numerical results are given, which involve theacceleration efficiency and convergence behavior of both ILU (k) preconditioners andKrylov subspace methods for solving nonsymmetric linear system originated from onekind of elliptic partial differential equation (PDE).
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(3)
32. 计算任意维几何体的近似中心
傅清祥,CheeK.Yap
数值计算与计算机应用    1999, 20 (4): 266-273.   DOI: 10.12288/szjs.1999.4.266
摘要725)      PDF(pc) (224KB)(252)    收藏
In this paper, an algorithm for computing near-centers of a nearly round body inany dimension is shown. The algorithm generalizes a recent result [2]. It is quite simpleand needs only d "parallel" probe steps including d(d + 1) probes in d-dimension.
参考文献 | 相关文章 | 多维度评价
33. 在PVM应用程序中调用ScaLAPACK库函数方法
张云泉,迟学斌
数值计算与计算机应用    1999, 20 (4): 274-282.   DOI: 10.12288/szjs.1999.4.274
摘要853)      PDF(pc) (421KB)(269)    收藏
As the parallel version of LAPACK, ScaLAPACK will provide convenient and powerful parallel programming platform for parallel computing society. In this paper, wepropose a PVM based ScaLAPACK application programming structure. Based on thisgeneral programming structure, we give three initialization methods of BLACS thatcan be used under different situations. The data distribution fashion used in ScaLAPACK is also introduced to help the user to understand them better. For those whoare familiar with PVM parallel programming, they can call ScaLAPACK subroutinesdirectly using our methods, without concerning ScaLAPACK details too much. Finally,several shortcomings of current version ScaLAPACK package are discussed and an APIis proposed to improve its usability.
参考文献 | 相关文章 | 多维度评价
34. 进气道跨音速流场的计算
欧阳洁,叶昌业
数值计算与计算机应用    1999, 20 (4): 283-292.   DOI: 10.12288/szjs.1999.4.283
摘要762)      PDF(pc) (346KB)(279)    收藏
In this paper, the implicit approximate factorization schemes is combined with themultigrid method for computing the transonic flow fields of A-27 inlet. The numericalresults have shown that the multigrid method based on the approximate factorizationschemes can eliminate numerical vibration, simplify parametric choice and accelerateconvergent rate.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(3)
35. 通过符号计算准确实现矩阵的Lanczos过程
张知难,信学工,张建宁
数值计算与计算机应用    1999, 20 (4): 293-301.   DOI: 10.12288/szjs.1999.4.293
摘要758)      PDF(pc) (264KB)(258)    收藏
In this paper we expound the principle why the Lanczos procedure of a matrix canbe accurately performed by symbolic computation and give the uses of the accuratelanczos procedure in obtaining the invariant factors of a matrix and solving the matrixequation AX-XB=C.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(8) CSCD(3)
36. 指数函数拟合曲线的最优基法
肖华勇,田铮,孙进才
数值计算与计算机应用    1999, 20 (4): 302-311.   DOI: 10.12288/szjs.1999.4.302
摘要847)      PDF(pc) (298KB)(290)    收藏
In this paper, an algorithm how to determine the optimal base functions and it'snumber is presented in fitting curve with exponential function. The method avoidselecting initial value and getting local extremum in fitting curve with nonlinear leastsquare method (NLSM). Numerical results show that the algorithm is very effective infitting of curve with exponential function.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(7)
37. 数值代数中值得探索的几个问题
蔡大用
数值计算与计算机应用    1999, 20 (4): 312-316.   DOI: 10.12288/szjs.1999.4.312
摘要802)      PDF(pc) (193KB)(330)    收藏
Based upon the author's research experiences, several topics in numerical algebraare proposed in this paper. Iterative solution for huge linear systems, theory of precondition, the relationship between pseudo-spectrum and behaviors of iterative algorithmsare emphased.Specially, l the author thinks of that the theory and algorithms for simultaneouspolynomial equations should be push forward in the near further.
相关文章 | 多维度评价
被引次数: Baidu(1)