1.This accelerating method can be extended to many other known methods such as Newton like methods,m step discrete Newton method and Brent method so that their convergence and stability are improved.
此方法具有一定的广泛性,可以直接推广到许多已有的方法,如拟Newton法、M步离散Newton法及Bren t方法等,均可在一定程度上改善这些方法的收敛速度和稳定性.
2.Secondly, we absorb the preconditioning idea of Davidson method and preconditioned Lanczos method and propose preconditioned inexact Newton method and preconditioned block inexact Newton method.
其次借鉴Davidson方法和预处理Lanczos方法的预处理思想，提出了预处理不精确Newton方法和预处理块不精确Newton方法。
3.In the paper we study the relationship between Davidson method and Newton method and emphasize on studying the inexact Newton method.
本文研究了Davidson方法与Newton方法的关系，并重点研究不精确Newton方法。
4.A new numerical method for solving two phase air water flow equations is developed The two phase air water equations are solved with a finite element method A discrete Newton iterative method is adopted to linearize the non linear finite element equations after temporal and spatial discretion, it not only converges faster than the other methods, but also avoids the difficulties on programming The mean of paralleling is adopted to solve the two phase finite element equations, reducing much EMS memory needed in calculation by integrating means
在利用有限元方法求解水气二相流方程时 ,引入了离散Newton迭代方法 ,用于非线性有限元方程组的线性化处理 ,将这一步计算的收敛阶由原有研究的线性收敛提高到平方收敛 ,并避免了直接应用Newton迭代方法给编程带来的不便。 同时在求解两相的有限元方程组时 ,采用两相方程组并行迭代的方法 ,与联立计算相比节省了大量的内存空间。
5.By reviewing the interpolation algorithm in ancient China (also in India and Arabia) and comparing this of Newton and Lagrange, it is obvious that orient algorithm had to be completed using the characteristics of the practical problem, so it was difficult to breakthrough generally. But the problem in the face of Newton and Lagrange was of universality so they succeeded in getting a general form.
二、考察中国古代、印度、阿拉伯、日本的插值算法的发展历史，对比西方Newton、Lagrange等插值算法的来源与成果，认为东方插值的实际背景独特，需要借助实际问题的特点来实现，因而难于作一般性的突破，而牛顿面对的问题却是有一般性的纯粹数学形式的问题，得到的也就是一般性的插值公式。
6.On the basis of Newton Kantorovich theorem, this paper gives the Kantorovich-type analysis for two quasi-Newton method: DFP and BFGS, using the technique given in [4].
本文在Newton－Kantorovich定理基础上，用［4］冲技巧对两种拟牛顿法－DFP和BFGS方法给出了Kantorovich分析．
7.Considering of the above algorithm is not appropriate to be realized in DSP (Digital Signal Processor), three additional location algorithms are presented: the improved Gauss-Newton iterative algorithm, the intersectant circle chord, the least square estimation.
针对常规Gauss-Newton迭代定位算法不易于在DSP(Digital Signal Processor,数字信号处理器)中实现的缺点,介绍了三种定位算法:改变目标函数的Gauss-Newton迭代定位算法、相交圆弦线方法和最小二乘估计算法.
8.Application of ANN to short-term power load forecasting basedon simulated annealing Gauss-Newton algorithm
基于模拟退火的Gauss-Newton算法神经网络在短期负荷预测中的应用
9.The Minimizer of a Certain Functional in Newton Space Belongs to the De Giorgi Class
Newton空间中某类泛函极小属于De Giorgi类
10.THE IMPROVED NEWTON-EULER DYNAMICS MODEL OF 6-DOF PARALLEL ROBOT
改进的6-DOF并联机器人Newton-Euler动力学模型
