近似方法

The main idea of the finite difference method is as follows: We consider the discrete difference equation with the finite alphas as approximate substitution for the differential equation with the continuous variant and boundary conditions.

有限差分方法的基本思想是用离散的，只含有有限个未知数的差分方程去近似代替连续变量的微分方程及边界条件，并把相应的差分方程解作为微分方程的近似解。

The basic concept in these sections is replacing backward Euler's method by modified backward Eulers's method to approximate the characteristic differential equation and discretizing the convection-diffusion equation at the middle point of the approximate characteristic curve, which will make the convective term get up to second order accuracy.

这三节的基本离散方法是用改进的Euler公式代替向后Euler公式来近似特征微分方程，并且将近似特征线的中点作为离散节点，使得对流项的离散精度提高到二阶，对于扩散项，将其转移到待求时间层进行离散。

For the design of the sliding mode control of time-delay large-scale systems, we propose the adaptive fuzzy sliding mode control schemes, and the fuzzy approximator is used to approximate the upper bound of the uncertainties which combine time-delay interconnections with perturbations in the system.

所以，我们针对具时间延迟的大型系统的滑动模式控制设计上提出了一个适应模糊滑动模式控制方法，其中，模糊近似器被用来近似系统当中时间延迟以及扰动相互结合后的上界。

It enlarges the range of quasi-linear Born approximation, and makes the prestack depth migration method based on quasi-linear Born approximation to be able to adapt strong lateral velocity variation.

在散射波场计算中，使用了一个更稳定的散射波场计算公式，扩大了拟线性Born近似的应用范围，使基于拟线性Born近似的叠前深度偏移方法能够适应更强的横向速度变化。

And the effects of length of data N to the approximate entropy of chaotic system isalso analyzed.All these are accomplished by examining several time series produced by a surperposition of white noise and noise-free data.It uses the discrete system logistic map and continuous system Lorenz attractoras the examples.

以Logistic方程描述的离散系统和Lorenz方程描述的连续系统为例，采用白噪声和无噪声信号以不同的信噪比混迭而成数据的方法，分析了相同数据长度N，不同信噪比下，混沌系统近似熵的变化；相同信噪比，不同数据长度对系统近似熵的影响。

Because of the problem of convergence, the key techniques to complete the one step simulation were explored thoroughly. The geometry mapping method based on finite element model and inverse deformation was developed to get the approximate initial blank for one step simulation. The effective selection of convergence factor in Newton-Raphson method was studied. The divergence criterion in iterative process for one step simulation was put forward. Also the countermeasure to ill equations coefficient matrix was studied.

由于存在收敛性的问题，本文重点研究了一步模拟方法计算机实现的关键技术：提出了基于有限元模型的几何映射方法和反向变形方法来确定初始毛坯的近似解，研究了带收敛因子的Newton-Raphson方法中如何高效地选取收敛因子，提出了一步模拟迭代求解过程中的发散准则，并研究了病态方程组系数矩阵的修正方法。

From 1920s up to Now, various techniques and methods for wave field decomposition, propagation and migration/imaging have been well developed, such as the Kirchhoff asymptotic method, frequency-wavenumber domain phase-shift and phase-shift-plus-interpolation methods, and the one-way wave equation based phase-screen and generalized screen methods, etc.

自上个世纪二十年代至今，波场分解、传播与偏移成像技术经历了将近一个世纪的发展，形成了多种不同的方法，如Kirchhoff波动方程高频近似解方法，频率-波数域的相移(phase-shift)方法和相移-内插方法，以及在混合域中基于单向波动方程的相位屏、广义屏方法等。

Results show that radial basis function and point interpolation methods possess Kronecker Delta function property, but the robustness is poor in some cases; the MLS approximation function does not possess Kronecker Delta function property, but it has good robustness. Differences among the three discretization schemes of meshless method are as follows:the collocation method requires no numerical integration and very little computational time while its robustness is poor; Galerkin method is not a truly meshless method due to the background meshes required for integration; the Petrov-Galerkin method is a truly meshless method and need numerical integration in each sub-domain, so it needs more computational time.

分析结果显示：径向基函数和点插值法均具有d 函数属性，但计算稳定性差；移动最小二乘近似函数不具有d 函数属性，但计算比较稳定；无网格方法中的3种离散方法不同之处在于：配点法不需要积分，计算量小，计算稳定性差；Galerkin方法需要借助背景网格进行积分，它不是真正的无网格方法；Petrov-Galerkin方法，是一种真正的无网格方法，它需要对每个子域进行积分，计算工作量较大。

The full text is total to is divided into four chapter:Chapter 1, the relevant concept of the listed company results evaluation, that is the other the theories foundation of the chapter, and do the foul to the textual evaluation angle settle;Chapter 2, the results evaluation index sign and the index sign system, the original index sign system of comprehensive evaluation of information formation that that choice listed company publish, and introduced the in common use index sign to sieve to endow with the method of the power with index sign;Chapter 3, the comprehensive results evaluation method, this is the foundation that carries on chapter 4, that carry on theories in chapter 2 fusion, became two kinds of comprehensive evaluation methods, and carried on the des cription to the operation step of a method;Chapter 4, the substantial evidence is analytical, this is the point of the full text, the collections combines the data of tidy up the listed company, making use of two kinds of methods within chapter 3s respectively, carry on synthesize to rank, and get two kinds of methods to rank the result to look like the consistent conclusion.

全文共分四章：第一章，上市公司绩效评价的有关概念，该章是其它章节的理论基础，并对本文的评价角度做出界定；第二章，绩效评价指标及指标体系，该章选择上市公司披露的信息形成综合评价的原始指标体系，并介绍了常用的指标筛选和指标赋权的方法；第三章，综合绩效评价方法，这是进行第四章的基础，该章将第二章中的理论进行融合，形成了两种综合评价方法，并对每种方法的操作步骤进行了描述；第四章，实证分析，这是全文的重点，收集并整理上市公司的数据资料，分别利用第三章中的两种方法，进行综合排名，并得到两种方法排名结果近似一致的结论。

In order to know irradiant dynamic process we must have excitation crosssection of electron collision with molecule. Incident electron energy is inintermediate range. At low energy, there are several good methods thatcalculate electron collision with molecule for example Variation, R-matrix.

此时入射电子的能量已经达到中能，在低能范围内的电子与分子散射有许多好的方法如变分法、R-矩阵方法，在高能范围内利用Born近似等方法也能得到满意的结果，在中等能量范围的电子与分子碰撞则没有相应的成熟的方法，使用比较多的是密耦(Closing-Couple)方法，这种方法将相互作用势分为静态势、交换势、极化势，在耦合态的选取上比较自由，可以根据实际的问题及分子的特性选取耦合作用比较强的通道，这样既能保证精度又能减少计算量。

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。