算法的

The main works include five parts:(1) A modified K-means algorithm for optimizing the RBPNNs' structures was proposed;(2) The RBPNNs' structure optimization method based on the orthogonal least square algorithm was proposed that can greatly reduce the computation expense for the structure optimization;(3) Using the GA to perform the whole structure optimization of the RBPNNs was proposed in this thesis, which include simultaneous optimization of the hidden centers and the corresponding controlling parameters of the kernel functions.

主要工作体现在如下三个方面：(1)提出一种改进的 K-Means 算法；(2)提出了基于递推正交最小二乘算法的结构优化算法来训练径向基概率神经网络，从而大大减小了用于结构优化的计算开销；(3)提出使用遗传算法来实现径向基概率神经网络的全结构优化，即隐中心矢量和核函数控制参数同时优化，通过新设计的编码方式、新构造的适应度函数，充分发挥了 GA 的全局搜索性能，使得所优化的径向基概率神经网络的结构趋于最简。

The study establishes the compression of models and algorithms for vector graphics data on different basis, including on the basis of vector graphic factors of B spline curves in predecessors research; on the basis of rectangle graphic factors in main axis mode; on the basis of the application of Hough transform to polygonal line, ellipse and circle, so the study improves the models and algorithms of compression for vector graphic data.The compression for vector graphic data mainly studies the data of storing graphic factors.

本文研究的内容是：根据对以往各种曲线矢量数据压缩模型与算法的研究，建立了基于B样条的曲线矢量图形要素的数据压缩模型和算法；在前人研究的基础上，建立基于主轴模式的矩形图形要素的数据压缩模型和算法；本文作者将Hough变换应用到折线、圆及椭圆图形要素的矢量数据压缩上，在此基础上改进了基于Hough变换的折线、圆及椭圆图形要素的数据压缩模型和算法。

The projection gradient method will be a possible way to solve the problem that we just get. It has been shown that the projections of the every directions, of which is the boundary point in linear restraint problems, are the possible decent directions, and the projection of negative grads direction is a decent direction. In 1960, Rosen proposed the basic idea of projection gradient methods, and then lots of researchers have been tried to find the convergence of this method. But most of them get the convergence with the condition to amend the convergence itself.

在约束最优化问题的算法中怎样寻找有效的下降方向是构造算法的重要内容，在寻找下降方向方面可行方向法中的投影梯度法有效的解决了下降方向的寻找问题，利用线性约束问题边界点的任意方向在边界上的投影都是可行方向，而负梯度方向的投影就是一个下降方向。60年代初Rosen提出投影梯度法的基本思想，自从Rosen提出该方法以后，对它的收敛性问题不少人进行了研究，但一般都是对算法作出某些修正后才能证明其收敛的，直到最近对Rosen算法本身的收敛性的证明才予以解决。

Finally, the programme of design for routing protocols is chose in the thesis. Secondly, improved algorithms are introduced in detail. In order to achieve to reduce duplication of data sent in, save energy, prolong the network lifetime and balance the network load, improved algorithms are based on the minimum hops algorithm which is one of the energy-aware routing, When the characters of the minimum hops algorithm are analysed in detail,multi-path routing, adaptive dynamic routing, routing reliability are designed.Then a new algorithm based on minimum hops in plane routing protocol and a new algorithm based on minimum hops in clustering routing protocol are designed in the thesis.

接著详细介绍了本文的改进路由算法，为实现减少数据的重复发送量、节约能量、延长网络寿命、平衡网络负载的目标，本文以能量感知路由之一的最小跳数算法为基础，详细分析了最小跳数算法的特点，设计了多路径修正路由，自适应动态路由，路由可靠性机制三方面的改进方案，并且结合改进方案设计了基于最小跳数的改进平面路由协议和基于最小跳数的改进分簇路由协议，将两种类型算法的优点相结合，提高网络的健壮性与可扩展性。

In chapter 4, basing theories and methods of scientific visualization, and artificial neural network BP algorithm, we integrate the Visual C++, OpenGL graphics library and Excel VBA technique to develop the program of artificial neural network and to make the BP algorithm visually, this program works can be divided into four parts: Using C language to develop program about BP algorithm; Using Visual C++, develop the GUI Interface, make input parameter visually; Using OpenGL graphic technique to display the training sample point in three dimension; at last using Excel DDE technique display the error graphic tables in Excel system In chapter 5, on the view of engineering application, we establish new method of surface reconstruction basing artificial neural network, develop interface program between module and commercial CAD/CAM system, meantime deeply discuss some key problems, for example, setting up the base plane, using the API technique, cutting and editing surface boundary, and also discuss the more compliant problem: how to intersect surface, at end we finish the work of translation from our surface reconstruction module to commercial CAD/CAM system, then make reverse engineering system basing artificial neural network more useful.

第四章基于科学计算可视化理论，依据人工神经网络BP算法理论模型，综合Visual C++，OpenGL图形库以及Excel VBA等多项软件开发技术，编制了人工神经网络程序，实现了BP算法的可视化映射。具体工作分为四部分：利用C语言实现人工神经网络BP算法；利用VisualC++的GUI技术开发图形用户界面，实现参数设置可视化；利用OpenGL图形技术进行三维映射，显示学习样本及训练样本点；利用微软电子表格DDE动态数据交换技术，在Excel上动态显示学习误差曲线图。第五章从工程应用的角度出发，提出了一种新的基于人工神经网络算法的曲面裁剪重构方法，完成了曲面重建模块与通用CAD/CAM系统的接口设计工作，对其中的若干关键问题进行了深入讨论，例如基平面设定、API技术的应用、边界裁剪等问题，同时，对曲面计算中较为困难的曲面相交问题也进行的专门探讨，最终完成了曲面重建模块向CAD/CAM系统的数据传输工作，使人工神经网络逆向工程系统趋向实用。

When processing the restrained conditions, we convert the restrained conditions into punish function and add it to fitness function which profits from the conventional method. We introduce the support vector machine to the selection operator, which increases the multiplicity of selection and increases the probability of the algorithm getting the overall optimal solution. Also, we use the adaptive probabilities of crossover and mutation in mixed GAs which enhances the efficiency of the algorithm. The result of tests about the system indicates that the improved genetic algorithm has more high convergence speed and better solution quality.

在生成初始种群时，不是简单的随机生成，而是用支持向量机对历史数据进行分类，用较优的一类历史数据作为初始种群；在约束条件的处理上，本文借鉴常规方法中的罚函数法，采用将约束条件作为罚函数包含到适应度评价中去的方法；在选择算子上，本文引入了支持向量机的方法，增加了选择的多样性，增加算法搜索到全局最优解的概率；同时针对交叉概率和变异概率在算法的前后期不同的情况，本文引入了Srinvivas等提出的动态自适应的方法，使得交叉概率和变异概率能够随着算法中个体的质量动态改变，提高算法的效率。

There exits several MDO algorithms. But they are in our opinion either nonefficient or complicated. So we devised an algorithm called Subspace Approximation Optimization . In the SAO algorithm, the whole system is decomposed into one system-level optimization and several disciplinary optimizations so that a large and complicated problem can be divided into several easy-solving sub-problems. The coupling relationships and the coordination among disciplines are presented by equality constraints and these equality constraints are assigned to relevant disciplines. The optimums of design variables in system level optimization are transferred to discipline level optimization. The optimums of design variables in discipline level correspond to the point that is the nearest to the optimums of design variables in system level. If the optimums of design variables in system level are out of feasible region of discipline 1eve1, linear constraints can be built in the system level optimization using the design variable optimums obtained by the discipline level optimization. The system level optimization would improve the design of the whole system with these linear constraints.

目前，国内外已经发展出了多种飞机多学科设计优化算法，本文的重点是针对协同优化算法的不足，提出了子空间近似优化算法(Subspace ApproximatingOptimization，SAO)，SAO算法中，整个系统的优化问题被分解成一个系统级优化和若干学科级优化，而各个学科之间的耦合与权衡关系则被当作等式约束，这些等式约束将被分配到各个学科级优化中去，系统级优化的任务是寻找整个系统的最优解，而学科级优化的目标函数是以系统级优化分配下来的设计点为圆心的超球半径的平方，因此，如果系统级优化分配下来的设计点在学科级优化可行域内，则学科级优化目标函数为0，反之，则学科级优化的最优点是系统级优化当前设计点距离可行域最近的点。

The main calculation in the particle image matching algorithm is to have the two visual seriate images taken the cross-correlation arithmetic based on the fast fourier transform. This paper discusses the basic principle of the cross-correlation arithmetic in detail, discuss each stage of its produce and development and also discuss the thinking, the merit and the defect of each algorithm. On the base of all these discussing, we confirm our algorithm and realize it in programming with C++.

粒子图像的匹配算法的主要运算是将两连续的视频图像进行基于快速傅里叶变换的互相关运算，本文详细论述了相关算法的基本原理，它的产生和发展的各个阶段，各种算法的思路和优缺点，在此基础上确立了自己的算法并以C++编程实现。

The chief points of this paper are summered as follows: Firstly, the dynamic rule-adjusting algorithms in Snort is analyzed. Secondly, some methods to improve the algorithms are discussed and the results of experiment are obtained. Thirdly, a kind of anomaly detection models using association rule mining algorithm is put forward, and the improvements of Apriori algorithm are pointed out. Fourthly, the advantage of the new algorithm and the feasibility of this model are testified. Lastly, a framework of distributed intrusion detection system based on data mining is designed, and distributed cooperation concepts and many detecting methods are adopted in order to guarantee the accuracy and completeness of the detecting system.

论文的主要工作如下：本文分析了Snort中的动态规则调整算法，讨论了一些改进的思路，给出了实验结果；提出了一种运用关联规则挖掘算法进行异常检测的模型，根据入侵检测的实际情况对Apriori算法进行了改进，通过实验证实了改进算法的优越性和模型的可行性；设计了一种基于数据挖掘的分布式入侵检测系统框架，采用分布式协作思想，结合多种检测方法，保证了系统检测的准确性和完备性。

On the basis of comparative analysis of the relevant micro-imagemosaicing algorithms, a suitable algorithm is choosed to combine local images to ahigh-resolution images of greater vision, which is better for pathologist to observe cervical cellmorphology. By some measurement algorithms of pathological cells and metal particulate which are reported in magazines, cervical cell parameters measuring and nucleus plasm ratio analysisare better realized.

本论文通过将成熟的图像处理算法有针对性地用于宫颈细胞图像处理，较好的实现了对宫颈细胞图像几何变换，色彩变换，平滑，增强，边缘检测；在对比分析相关显微图像拼接算法的基础上，选择合适算法将局部图像合成更大视野的高分辨率图像，供医生更好的观察细胞形态；将相关文献中报导的处理病理细胞和金属微粒形态测量的算法用于宫颈细胞形态参数的测量，较好的实现了细胞形态参数测量及细胞核浆比分析等功能。

But we don't care about Battlegrounds.

但我们并不在乎沙场中的显露。

Ah! don't mention it, the butcher's shop is a horror.

啊！不用提了。提到肉，真是糟透了。