中国科学院数学与系统科学研究院期刊网

计算数学 1999年 21卷

Please wait a minute...
选择: 显示/隐藏图片
1. 半无限规划问题的一个有效解法
周广路,王长钰,张玉忠
计算数学    1999, 21 (1): 1-8.   DOI: 10.12286/jssx.1999.1.1
摘要1150)      PDF(pc) (270KB)(665)    收藏
This paper presents an efficient method for solving unconstrained semi-infiniteminimax problems and constrained semi-infinite minimax problems and constrainedsemi-infinite problems, in which these two kinds of problems are approximated byunconstrained differentible optimization problems. Numerical examples are givento show the high efficiency of the method.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(39) CSCD(4)
2. 关于多重网格并行计算中拟边界Jacobi成份的影响
莫则尧!北京8009信箱计算物理实验室,李晓梅!北京怀柔3380信箱90分箱
计算数学    1999, 21 (1): 9-18.   DOI: 10.12286/jssx.1999.1.9
摘要1073)      PDF(pc) (539KB)(586)    收藏
During parallel multigrid computations, the Jocobi components along pseudoboundaries in the block by block relaxations operator become the main source todecrease the numerical efficiency of multigrid. We investigate these decrements indetail for constant coefficients Symmetric Positive Definite elliptic model problemsby error character1stic functions in this paper, then some strategies are gained. Atlast, advection-diffusion problems are addressed.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(7) CSCD(2)
3. 一类具有小周期系数的椭圆型边值问题的双尺度渐近分析方法
崔俊芝,曹礼群
计算数学    1999, 21 (1): 19-28.   DOI: 10.12286/jssx.1999.1.19
摘要1085)      PDF(pc) (333KB)(753)    收藏
A new type of computable scheme is provided for a class of elliptic bounaryvalue problems with small periodic coefficients. The principle idea of this methodis to change the computation of original problems into the solving process of theperiodic solution defined in the basic condguration and boundary layer. In thispaper, a completely rigorous mathematical theory for this process is presented.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(122) CSCD(4)
4. Navier-Stokes方程的最佳有限元非线性Galerkin算法
何银年,李开泰
计算数学    1999, 21 (1): 29-38.   DOI: 10.12286/jssx.1999.1.29
摘要1347)      PDF(pc) (369KB)(739)    收藏
A optimum ffote element nonlinear Galerkin algorithm is presented for thetwo-dimensional nonstationary Navier-Stokes equations. The standard finite elemellt Galerkin algorithIn consists in solving a nonlinear equation on the fine gridfinite elemellt space Xh' The optimum finite element nonlinear Galerkin algorithm consists in solving a nonlinear subproblem on a coarse grid finite elementspace XH(H > h) and solving a linear subproblem on a fine grid incremental finite element space Wh = (I - RH)Xh- If H is chosen such that H = O(h1/2),then two algorithms are of the c0nvergence rate of same order. However, sinceH >> h, dimXH << dimXh, the optimum finite elemellt nonlinear Galerkin algorithm can save a large amoullt of comPutational time. Finally, we give thenumerical test which shows the correctness of theoretical analysis.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(7)
5. 关于几种三角形板元等价性的注记
石钟慈,王家城
计算数学    1999, 21 (1): 39-44.   DOI: 10.12286/jssx.1999.1.39
摘要1115)      PDF(pc) (212KB)(597)    收藏
In this note, we point out that two generalized conforming triangular plate elements, namely GCⅢ-T9 and GCⅡ-T9, proposed in references [2] and [4], are identical to the familiar Specht element[3]. Moreover, another generalized conformingelement in reference [6] is proved to be equivalent to VZ1 element in [8].
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(10)
6. 带溶质吸附的混溶驱动问题的Galerkin方法
程爱杰
计算数学    1999, 21 (1): 45-58.   DOI: 10.12286/jssx.1999.1.45
摘要1160)      PDF(pc) (508KB)(533)    收藏
A priori error estimates for Galerkin method for numerical approximation ofthe coupled quasilinear system for c = c(x, t), p = p(x, t), cr = c.(x, t) given byfor x E fl,t E (O,T] are considered. Equations of this type arise in model forthe miscible displacement of one compressible fiuid by another with adsorption ofor orsolute in a porous medium. The cofficient D, which is a function of x, c, X -reflects moleculer diffusion and dispersion. Colltinuous-time Galerkin procedureis analysed. A new technique is illtroduced to establish the evolution inequlity oferror. The optimal error estimates in Lco(J; H1(Ω)) are presented.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(5)
7. 利用通量限制的高分辨Godunov型格式的熵条件
张虎,赵宁
计算数学    1999, 21 (1): 59-64.   DOI: 10.12286/jssx.1999.1.59
摘要1221)      PDF(pc) (194KB)(622)    收藏
In this paper, a high resolution Godunov scheme by using flux limiters is presented. The entropy inequality of the scheme is discussed here, which can result inthe convergence. And the constraint to the limiters is provided. Final conclusionshows that it is difficult for the scheme under the entropy condition to achievesecond-order resolution.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(6)
8. 预测校正公式求解指标2微分代数方程
曹阳,李庆扬
计算数学    1999, 21 (1): 65-74.   DOI: 10.12286/jssx.1999.1.65
摘要1197)      PDF(pc) (404KB)(696)    收藏
This paper combines the implicit multistep method and the half explicit multistep method to solve index2 differentiaLalgebraic equations (DAEs), proposesthe predictor-corrector formula. This method enlarge the set of multistep Inthodssuitable to solve index-2 DAEs and improve the mboum order of multistep methodfor solving index2 DAEs. This paPer discuss the global convergence and the im-plemellt of the method. Numerical test are also listed which show the method itproposed it better than BDF method when solving nonstiff DAEs.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(1)
9. 双障碍问题的逐次逼近阻尼牛顿法
马昌凤
计算数学    1999, 21 (1): 75-80.   DOI: 10.12286/jssx.1999.1.75
摘要1105)      PDF(pc) (205KB)(617)    收藏
In this paper, we present a successive approximation damped Newton methodfor bi-obstacle problems based on its equivalent nonsmooth equations. Under suitable conditions, we get the global convergence of the algorithm. Some numericalresults are also reported in the paper.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(5) CSCD(1)
10. 关于多点源势函数的快速计算
孙丰荣,田发中
计算数学    1999, 21 (1): 81-88.   DOI: 10.12286/jssx.1999.1.81
摘要1176)      PDF(pc) (346KB)(730)    收藏
The possible approaches of the reducibility of the refinemeot included in theFMM, and of the reduction of the amount of the work done by the algorithm arediscussed in this paper, based on the analyses of the methods of the refinement andthe compexity in various computing envoirmenis. The theoretical analysis and thenumerical examples indicate that the FMM-1 or the FMM-2 is better than FMM-min the implement of the algorithm and the computing efficiency for N ~ 103.
参考文献 | 相关文章 | 多维度评价
11. 利用远场模式反演声波阻尼系数
王连堂
计算数学    1999, 21 (1): 89-98.   DOI: 10.12286/jssx.1999.1.89
摘要1142)      PDF(pc) (317KB)(694)    收藏
The inverse time-hormonic acoustic wave scattering problem we are consideringin this paper is to recover the impedance coefficient from the far field pattern ofthe scattered wave. A method is presellted and the convergence of this method isproven. Numericale examples are given showing that this method is both accurateand simple to use.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(31)
12. 分布式系统上并行矩阵乘法
吴建平,迟学斌
计算数学    1999, 21 (1): 99-8.   DOI: 10.12286/jssx.1999.1.99
摘要1206)      PDF(pc) (393KB)(778)    收藏
By begiwhng with the Cannon algorithm[1] and the double- direct ion dat amoving algoritlun[2] for processors arranged as a 2-D square mesh, we improvethese two algorithms to general 2-D mesh in this paper. We also consider theapplication of our algorithms in those operations which are similar to matrix multiplication. Using MPI parallel programming environment, we have obtained satisfactory performance on Dawning-1000.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(35) CSCD(6)
13. 经典分形集测度上估的计算机搜索Ⅰ──对典型例子Sierpinski垫片编码技术的剖析
王何宇
计算数学    1999, 21 (1): 109-116.   DOI: 10.12286/jssx.1999.1.109
摘要1181)      PDF(pc) (270KB)(605)    收藏
The upper estimation of Hausdorff measure for Sierpinski gasket has beengreatly hoprovd in Refs. [1, 2]. As the basis of these two works, the codingteclmique of the upper estimation seardsng for regular fractal sets on computerdiscusses thoroughly in this paper through a troical example of Sierpinski gasket.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(12) CSCD(2)
14. 区域分解界面预条件子构造的一般框架
胡齐芽,梁国平
计算数学    1999, 21 (1): 117-128.   DOI: 10.12286/jssx.1999.1.117
摘要1768)      收藏
In this paper we propose a unified framework to construct interface preconditioner for non-overlapping domain decomposition methods, and prove a generalresu1t to estimate condition number of preconditioned interface matrices. Basingon this, we analyse the essence of the known interface preconditioners, and construct a kind of interface preconditioner for hybrid element domain decompositionmethods.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(9) CSCD(2)
15. 广义KdV-Burgers方程长时间性态的谱方法
吕淑娟,张法勇
计算数学    1999, 21 (2): 129-138.   DOI: 10.12286/jssx.1999.2.129
摘要1161)      PDF(pc) (328KB)(770)    收藏
In this paperl we study fully discrete spectral method of Generalized KdVBurgrs eqution with peridic initial conditions. The exstence of A, respectively and d(A, ap) - 0 are obtained. Finilly the large time error estimation for fully discrete approtimation solution is provd.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(4)
16. 两类变时间步长的非线性Galerkin算法的稳定性
何银年,侯延仁
计算数学    1999, 21 (2): 139-156.   DOI: 10.12286/jssx.1999.2.139
摘要1205)      PDF(pc) (609KB)(743)    收藏
This paper reprents two types of fully discrete Galerkin algorithm and nonlinear Galerkin algoritlun with variable time steps for solving numerically nonlinear evolution equations, in which spatial discretization is made by spectral functions and finite elements; time is done by the Euler explicit difference scheme with the first order accuracy and two-step semi-implicit difference scheme with the second order accuracy. According to the stability analysis, we find that for the Euler difference scheme and two-step difference scheme on time discretization the stability of the fully discrete nonlinear Galerkin algorithms is superior to ones of the fully discrete Galerkin a-lgorithms. Finally, our numerical test also shows this fact.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(9) CSCD(1)
17. 病态线性代数方程组的一种刚性问题数值解法
吴新元
计算数学    1999, 21 (2): 157-162.   DOI: 10.12286/jssx.1999.2.157
摘要1526)      PDF(pc) (243KB)(975)    收藏
Basing on the relationship between ill-conditioned linear systems of algebraic equations and stiff systems of ordinary differelltial equations and using an explicit single step method with 2-order for stiff systems, the numerical solution of the initial value problems relative to the ill-conditioned linear systems of algebraic equations is obtained. So does the approkimation solution of the ill-conditioned linear systems of algebraic equations.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(29) CSCD(3)
18. 关于离散代数Riccati方程的扰动分析
刘新国,郭晓霞
计算数学    1999, 21 (2): 163-170.   DOI: 10.12286/jssx.1999.2.163
摘要1186)      PDF(pc) (260KB)(741)    收藏
By using some intrinsic relations between the discrete-time algebraic Riccati equation (DARE for short) and a symplectic matrix pencil, this paper deals with perturbation analysis of the DARE. The results show that the sensitivity of the unique synunetric positive semi-definite solution of the DARE depends on the sepa-ration of the stable and unstable sub-pencils.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(2)
19. 欧拉方程的Velicity表示和辛积分
陈旻
计算数学    1999, 21 (2): 171-180.   DOI: 10.12286/jssx.1999.2.171
摘要1256)      PDF(pc) (305KB)(737)    收藏
In this paPer the author studies Buttke's velicity reformulation of the incompressible Euler equation, and the symPlectic integration of the Hamiltonian system obtained by discretizing the vehicity equation.The author shows that he linearized velicity equation is hyperbolic only in the weak sense and she analyzes the characteristics of the linea-rized velicity equation for the variable coefficient case. The author will briefiy describe the symplectic schemes: one is the imPlicit midpoint scheme, and tanother a fourthorder implicit scheme. Also, the author shows a computational result obtained by redistributing the Lagrangian points at flxed times.
参考文献 | 相关文章 | 多维度评价
20. 离散形式的声阻抗反演问题的稳定性分析
郝现军,张关泉
计算数学    1999, 21 (2): 181-188.   DOI: 10.12286/jssx.1999.2.181
摘要1181)      PDF(pc) (259KB)(580)    收藏
In this paper, we consider the acoustic impedance inversion problem of one dimensional wave equations. from the difference scheme of one diemsional wave equations, we derive the positive property of impulsive response in frequency domain,and prove that the positivity is a sufficient condition for the stability of impedance inversion.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(5)
21. 再生核空间W_2~2[0,∞)中一类积分──微分方程精确解的表示
李云晖,崔明根
计算数学    1999, 21 (2): 189-198.   DOI: 10.12286/jssx.1999.2.189
摘要1145)      PDF(pc) (310KB)(683)    收藏
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(46) CSCD(5)
22. 抛物型初边值问题的有限元与边界积分耦合的离散化及其误差分析
杜其奎,余德浩
计算数学    1999, 21 (2): 199-208.   DOI: 10.12286/jssx.1999.2.199
摘要1172)      PDF(pc) (376KB)(916)    收藏
In this paper, based on the study of [1], the discretizations of the coupling of finite elemellt and boundary integral are presented to solve the initial boundary value problem of parabolic partial differential equation defined on an unbounded domain.The semi-discrete scheme and fully discrete scheme are given, and stability theorem and error estimates, which correspond to discrete scheme respectively,are obtained.Finally,the numerical example is provided,and numerical result shows that the method is feasible and effective.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(20) CSCD(1)
23. H~2(I)空间中的离散小波变换
吴勃英,何耀东
计算数学    1999, 21 (2): 209-214.   DOI: 10.12286/jssx.1999.2.209
摘要1334)      PDF(pc) (167KB)(831)    收藏
In this paper, a boundary scale function b(x) with a compact support is given,it improves a boundary scale function without the compact support in [1], then a multiresolution analysis (MRA) in (I) space is constructed and a discrete wavelet transform in H2(I) space which maps function's sample values into the function's wavelet coefficients is given, so that the wavelet interpolation function is obtained.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(13) CSCD(2)
24. 基于通量ENO形式的多分辨算法
练永生,王汝权
计算数学    1999, 21 (2): 215-224.   DOI: 10.12286/jssx.1999.2.215
摘要1117)      PDF(pc) (323KB)(662)    收藏
In this paper, we use the flux ENO scheme for the multiresolution scheme and simplify the original scheme based on the cell average ENO. As it is well known that the present scheme may will not be conservative at discontinuities, but numerical solutions are acceptable in practice.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(3)
25. 流体弹塑性介质中波动与断裂计算的二阶GODUNOV方法
汤寒松
计算数学    1999, 21 (2): 225-236.   DOI: 10.12286/jssx.1999.2.225
摘要1212)      PDF(pc) (450KB)(720)    收藏
A finite difference method is proposed for modeling wave motion and fracture phenomenon in hydro-elasto-plastic bodies. The method is a Godunov type approach,and it has second-order accuracy and captures shocks and fracture zones with high resolution. In the class of given wav systems the Riemann problems involved in the method have unique solutions, and the solutions can be obtained with efficient procedures. Numerical results are satisfactory in computations of Riemann problems and spallation in a steel plate.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(1)
26. 弹性接触问题的一种新的混合变分形式
王烈衡,王光辉
计算数学    1999, 21 (2): 237-244.   DOI: 10.12286/jssx.1999.2.237
摘要1181)      PDF(pc) (234KB)(631)    收藏
In this paper, for the colltact problem in elasticity, we proposed a new mixed variational formulation, which is the base for the dual mixed finite element method of the contact problem.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(28) CSCD(2)
27. 矩形域与三角域Bezier曲面间的PPG~n变换
江雷,王仁宏
计算数学    1999, 21 (2): 245-250.   DOI: 10.12286/jssx.1999.2.245
摘要1232)      收藏
Geometric continuity is a very important issue in CAGD and has gained more attention. In this paper we present the new concept of polynomial preserving Gn cootinuity (PPG") between rectangular and triangular Bezier surfaces. Necessary and swhcieot conditions of PPG" are established explicitly. A simple and useful sufficient condition is also presented. The results can be used more effectively in connecting surfaces and composing rectangular and triangular patches.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(1)
28. 旋转Q1非协调元的V循环多重网格法
许学军
计算数学    1999, 21 (2): 251-256.   DOI: 10.12286/jssx.1999.2.251
摘要1217)      收藏
In this paper, an optimal V-cycle muligrid method is presented for Q1 nonconforming element. The uniform convergence rate independent of mesh size and level is established.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(11) CSCD(2)
29. Burgers方程的混合元分析及其数值模拟
罗振东,刘儒勋
计算数学    1999, 21 (3): 257-268.   DOI: 10.12286/jssx.1999.3.257
摘要1868)      PDF(pc) (429KB)(930)    收藏
In this paper, the 1-D Burgers equation is studied, the existence of its mixed generalized solution and semi-discrete and fully discrete mixed finite element solu-tions are discussed, and their error estimates are analyzed. And finally, an example of numerical simulation of fully discrete mixed finite element solution is also given.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(83) CSCD(8)
30. N-S方程一般近似惯性流形构造和逼近
李开泰,侯延仁
计算数学    1999, 21 (3): 269-282.   DOI: 10.12286/jssx.1999.3.269
摘要1268)      PDF(pc) (482KB)(750)    收藏
In this paper, three general principles for constructing approximate inertial manifolds are provided, under which the associate approximate inertial form of origin problem, which is a finite dimensional ordinary differential equation, is well-possed and its solution will approximate the genuine solution at some degree. At last, for some kinds of approximate inertial manifolds and a family of approximate inertial manifolds, we indicate that the principles given here are suitable.
相关文章 | 多维度评价
被引次数: Baidu(5) CSCD(1)
31. 非自伴椭圆问题的离散强极值原理与区域分解法
胡健伟
计算数学    1999, 21 (3): 283-292.   DOI: 10.12286/jssx.1999.3.283
摘要1173)      PDF(pc) (442KB)(750)    收藏
The solution of the boundary-vaue problem for non-self-adjoint elliptic equa tions is approximated by Partial Upwind Finite Element method, where all the angles of the triangles a are /2 but the mesh parameter h are arbitrary and which insures the validity of the strongly maximum principle for the discrete prob-lem. The Schwarz alternating method will enable us to break the discrete linear system into several linear subsystems of smaller size and we shall show that the approximate solutions from Schwarz domain decomposition method converge to the exact solution of the linear system geometrically and uniformly.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(6) CSCD(1)
32. 可显式求解二维扩散方程组的三层差分格式
杨国强,陈光南
计算数学    1999, 21 (3): 293-302.   DOI: 10.12286/jssx.1999.3.293
摘要1207)      PDF(pc) (308KB)(709)    收藏
In order to resolve the two-dimensional diffusion system, a kind of simple and explicitly resolvable trilayer difference scheme is adopted in this paper. In case of Const, the unconditional stability and convergence under the H1 norm are proved. The convergence rate is .
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(3) CSCD(1)
33. 求解线性矩问题的一个修正方法
王连堂
计算数学    1999, 21 (3): 303-308.   DOI: 10.12286/jssx.1999.3.303
摘要1154)      PDF(pc) (193KB)(664)    收藏
In this paper a modified method for linear moment problem and the error estimate are presented. It is proven that the modified method is a kind of projection method under suitable conditions. Numericale examples are given showing that this method is both accurate and simple to use.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(16)
34. 连续化方法求解约束凸规划问题
林正华,于晓林,于波
计算数学    1999, 21 (3): 309-316.   DOI: 10.12286/jssx.1999.3.309
摘要1343)      PDF(pc) (352KB)(762)    收藏
In this paper, we present a homotopy continuation method for a class of large constraints convex programming problems. The first step, the constraints are ap-proximated by a family of smooth aggregate constraints, the second step, we con-struct a homotopy method for globally finding convex programming problems with one constraint.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(5)
35. 一种有限元亏量校正格式的误差估计
杨一都
计算数学    1999, 21 (3): 317-324.   DOI: 10.12286/jssx.1999.3.317
摘要1200)      PDF(pc) (277KB)(697)    收藏
In this paper, the error of the following defect correction process are discussed.where uh E S1h is the linear finite element solution and S. Under natu-ral smoothness assumption, it is proved that the correction solution I2uhincreases the accuracy for the Uh.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(1) CSCD(1)
36. 并行迭代算法的有效性的度量参数
白中治,高作峰,黄廷祝
计算数学    1999, 21 (3): 325-330.   DOI: 10.12286/jssx.1999.3.325
摘要1154)      PDF(pc) (206KB)(653)    收藏
For a class of ideal models of parallel computers, we define some measuring parameters such as the speed-up, the efficiency, the redundancy of a linear and nonlinear parallel iteration method in both average and asymptotic senses, as well as the utilization ratio of the parallel computer. These parameters are reasonable and convenient for the theoretical studies of the parallel iteration methods.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(6)
37. 复合材料拟周期结构的均匀化方法
曹礼群,崔俊芝
计算数学    1999, 21 (3): 331-344.   DOI: 10.12286/jssx.1999.3.331
摘要1137)      PDF(pc) (413KB)(864)    收藏
In this paper, we shall discuss the homogenization problem of boundary value problems for the systems of linear elasticity with the quasi-periodic microstruc-tures, and give several basic estimations for displacement,stress and strain en-ergy,which are the basis of finite element computing.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(103) CSCD(4)
38. 经典分形集测度上估的计算机搜索Ⅱ──对典型例子Sierpinski垫片计数技术和格点跟踪技术的剖析
王何宇,王兴华
计算数学    1999, 21 (3): 345-354.   DOI: 10.12286/jssx.1999.3.345
摘要1166)      PDF(pc) (362KB)(766)    收藏
A upper estimate function v(x) of Hausdorff measure H'(S) of Sierpinski Gas-ket is given. A mathematical representation of the upper approximate value uN(x)to u(x) and a simple algorithm of uN(x) based on lattice tracing technique are also derived. As a simple corollary, the estimation H'(S) min u15(n 10-5)=u15(0.50783)=0.81794'is obtained.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(13)
39. 三项共轭梯度法收敛性分析
戴彧虹,袁亚湘
计算数学    1999, 21 (3): 355-362.   DOI: 10.12286/jssx.1999.3.355
摘要1200)      PDF(pc) (309KB)(865)    收藏
This paper studies the three-term conjugate gradient method for unconstrained optimization. The method includes the classical (two-term) conjugate gradient method and the famous Beale-Powell restart algorithm as its special forms. Some mild conditions are given in this paper, which ensure the global convergence of general three-term conjugate gradient methods.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(32)
40. 一类非线性对流扩散问题的FDSD预测校正格式
张强,孙澈
计算数学    1999, 21 (3): 363-374.   DOI: 10.12286/jssx.1999.3.363
摘要1229)      PDF(pc) (470KB)(851)    收藏
In this paper, a predictor-corrector finite difference-streamline diffusion finite element scheme is constructed for time-dependent quasilinear convection-diffusion problem. For the scheme considered, the solvility is proved, and the error estimate in L(L2)-norm is established. It has 2-order accuracy in time direction and quasi-optimal order accuracy in space variables.
参考文献 | 相关文章 | 多维度评价
被引次数: Baidu(30) CSCD(2)