法的理论

Among used machine learning methods, the gradient descent method is widely used to train various classifiers, such as Back-propagation neural network and linear text classifier. However, the gradient descent method is easily trapped into a local minimum and slowly converges. Thus, this study presents a gradient forecasting search method based on prediction methods to enhance the performance of the gradient descent method in order to develop a more efficient and precise machine learning method for Web mining.However, a prediction method with few sample data items and precise forecasting ability is a key issue to the gradient forecasting search method. Applying statistic-based prediction methods to implement GFSM is unsuitable because they require a large number of data items to model a prediction model. In the contrast with statistic-based prediction methods, GM(1,1) grey prediction model does not need a large number of data items to build a prediction model, and it has low computational load. However, the original GM(1,1) grey prediction model uses a mathematical hypothesis and approximation to transform a continuous differential equation into a discrete difference equation in order to model a forecasting model.

其中梯度法是一个最常被使用来实现机器学习的方法之一，然而梯度法具有学习速度慢以及容易陷入局部最佳解的缺点，因此，本研究提出一个梯度预测搜寻法则(gradient forecasting search method， GFSM)来改善传统梯度法的缺点，用来提升一些以梯度学习法则为基础的分类器在资讯探勘上的效率与正确性；而一个所需资料量少、计算复杂度低且精确的预测模型是梯度预测搜寻法能否有效进行最佳解搜寻之关键因素，传统统计为基础之预测方法的缺点是需要较大量的数据进行预测，因此计算复杂度高，灰色预测模型具有建模资料少且计算复杂度低等优点，然而灰色预测理论以连续之微分方程式为基础，并且透过一些数学上的假设与近似，将连续之微分方程式转换成离散之差分方程式来对离散型资料进行建模及预测，这样的作法不尽合理，且缺乏数学理论上的完备性，因为在转换过程中已经造成建模上的误差，且建模过程仅考虑相邻的两个资料点关系，无法正确反应数列未来的变化趋势。

Based on Wilson\'s lemma on evenly distributed differences, elementary number theory and fundamental theory of finite field, the finite field construction for strict optimal OOCs is proposed and the basic idea of this construction is investigated. The main steps and flow chart of assistance design by computer are obtained.

3基于Wilson均匀分布差引理、初等数论及有限域的基本理论，提出一种严格最佳光正交码的有限域构造法，对该构造法的基本设计思想进行深入的理论分析，得到该构造法计算机辅助设计的主要步骤及流程。

The ice sensor is made of permanent flexibility alloy material which have magnetostriction ability.

国外在冰传感器的理论研究方面取得了很大进展，常用的有红外线法、电容法、热流法、超声法、光学法、微波法、振动法。

Parallel to the nomographic method, the numeri- cal calculating method is also put forward here. As an example, the lead and aluminium rings are pressed with oil lubrication or with- out. The results of data processing by nomogram and the numerical calculating are compared.

在线图法处理实验数据的基础上，提出了理论计算法，并对铅、铝两种材料在有无润滑剂条件下进行了圆环压缩试验，比较了用线图法和理论计算法对实验数据的处理结果，指出了各种方法的优缺点。

Chapter one is "Introduction". On the basis of summarizing the history of the study of the pictophonetic characters, we suppose some theoretical principle, such as the principle of the correspondence between the form and the meaning of the individual graph of Chinese characters, the principle of the total graphs of Chinese characters being a character-formation system, and the principle of the optimum development of the character -formation system in the history, and som analytic methods, such as the methods of structrue-function analysis, form-position analysis and meaning factor analysis, in the study of the pictophonetic characters. Chapter two is titled"General knwledge of the pictophonetic character." We discuss their origins, nature and some standards of distingnishing them, introduce the knowledge about the editions of SHUO WEN JIE ZI, analyse the functions of the modules of pictophonetic characters in detail, discuss concretely the function of phonetic graphs showing origins and pronunciations, the function of meaning graphs showing meanings and analogies, and the function of distinguishment when the two kinds of graphs are conbined. Chapter three is"The system of meaning graphs of Xiao Zhuan system."After we describe the structure of the meaning graph system, the relations between meaning graphs, the distributions of the function of meaning graphs showing meanings, and the frequency of all this kind of graphs, we get a table of all meaning graphs, a table of the frequency. Chapter four is"The system of phonetic graphs of Xiao Zhuan system."We describe the structure of the phonetic graph system, the relations between phonetic graphs, and between their function showing origins and their function showing pronunciations, and their frequency, then we get ageneral table of phonetic graphs and table of their frequency. Chapter five is "conclusion". We think that systematical methodology is useful in describing and analysing the formation system of Chinese characters. The methodology is adaptable in describing the formation system of modern Chinese characters as well as the historical character systems, adaptable in describing horizontally as well as comparing vertically. Only when we finish the systematical vertical compare on the base of systematical horizontal decription, could we know the true history of the system of Chinese characters, recover the rule of the system's development, and supply the theoretical supports for the scientifical arrangement of modern Chinese characters.

第一部分—绪论，在总结形声字研究历史状况的基础上，提出了研究形声系统的理论原则和分析方法，即汉字个体字符形义统一的思想，汉字字符集合构形系统的思想，汉字构形系统历史演变系统优化的思想等三项理论原则和结构功能分析法，形位分析法，义素分析法等三种分析方法，还扼要地介绍了具体的工作程序和论文的主要内容；第二部分—形声字概述，论述了形声字的来源、性质和鉴别标准，介绍了对《说文解字》正篆形声字说解中的传抄讹误，版本异文，说解错误校勘和考证的经过，详细分析了形声字构件的功能，具体讨论了声符的示源功能与示音功能，义符的示意功能与类化功能以及声符与义符组合中的相互区别功能；第三部分—小篆字系的义符系统，从义符系统的构成、义符间的相互关系，义符示意功能的分布及义符的构字频度等不同侧面，对义符系统进行了全面的描写，产生了义符总表、义符间意义关系总表和义符构字频度总表；第四部分—小篆字系形声字的声符系统，从声符系统的构成、声符间的相互关系，声符的示源功能与示音功能的关系以及声符的构字频度等不同侧面，对声符系统进行了全面描写，产生了声符总表和声符构字频度总表；第五部分—结束语，在全面测查小篆字系形声字的基础上，对研究汉字的方法和现代汉字发展方向的问题提出了一些看法，认为系统论的方法是描写和分析汉字构形系统行之有效的方法，系统论方法，不仅适用于历史上文字系统的描写，也适用于现代汉字构形系统的描写，不仅适用了断代描写，也适用于历史比较，只有在对汉字断代系统描写的基础上进行系统的历史比较，才能认清汉字系统历史演变的真相，揭示汉字系统发展的客观规律，为科学地整理现代汉字提供理论上的依据。

First, this article use the theoretical t-z curve to simulate the single pile's working status, compared with the elastic method, the layered soil's property can be better thought over.

首先在理论t—z法的基础上计算了单桩的p—s曲线，与弹性理论法相比，t—z曲线法能更好地考虑土的成层性。

Till now, there are a lot of generators such as the classical Fibonacci method, mid-square method, linear congruential method, feedback shift register method and combined method; the recent nonlinear congruential method, decimal fraction method, add-with-carry and substract-with-borrow methods, generalized feedback shift register method and MT method, and so on.

至今关于随机数发生器有很多的理论和方法，其中除了传统的Fibonacci法、平方取中法、线性同余法、位移寄存器法和组合方法；最近还有非线性同余法、取小数法、进位加和错位减法、广义反馈位移寄存器法和MT法等等。

The specific contents are as follows:Firstly, the causes of energy balance method result has some errors is analyzed. Then energy balance method is improved from the following aspects: the wind power input, the conductor self-damping, the damper dissipated power and vibration solution of Stockbridge dampers-conductor coupling system. So that improved energy balance method can consider more influential factors to enhance accuracy.

本文对输电线的微风振动控制和次档距振荡进行了多方面的研究，主要包括能量平衡法的改进、防振锤-输电线耦合系统的振动求解、阻尼线的耗能计算理论、微风振动的室内模拟试验、分裂导线微风振动的精细计算理论、多分裂导线的最大次档距和阻尼间隔棒优化布置等几部分，具体如下：首先分析了微风振动传统计算方法能量平衡法计算误差较大的原因，分别从风能输入功率、输电线自阻尼功率、防振锤消耗功率、输电线-防振锤耦合体系的振动求解等方面，对能量平衡法进行了改进，使能量平衡法可以考虑更多的影响因素以提高求解精度。

With the support of a major research grant of Zhejiang Province (No. 001101027) and the Plan of Promoting Education of Excellence of Zhejiang University in the 21st Century, this dissertation is devoted to address the above problems and fabricate an L-band EDFA with low cost, high performance and technological innovation.First the mechanism of an erbium doped fiber amplifier is introduced. Various theoretical models based on classical rate propagation equations are summarized and compared for their applications. We use a modified numerical model based on the Giles model, which agrees with the experimental results well. This serves as the theoretical base for the whole thesis and will be used to optimize the characteristics of EDFAs. A novel method based on a map of noise figure and gain is proposed to analyze efficiently the dependence of the noise figure and gain for EDFAs on the EDF length and the pump power. The characteristics of various EDFAs are studied and compared.Secondly, some important parameters are introduced to describe the characteristics of EDFAs. The methods and the experimental setups for measuring the gain and noise figure of an EDFA are emphasized. We introduce how to measure the characteristics of each component of an EDFA, especially the gain and absorption coefficient of an erbium doped fiber.Thirdly, four types of simple and novel L-band EDFA structures with low noise, high gain and low gain ripple are proposed to satisfy the requirement of a DWDM system. They can be classified into two categories. In the first category, a single ASE pumping is served asa secondary pump. This category includes the following three types. Type: a new construction using a fiber Bragg grating in an unpumped EDF section at the input side. Type: a new structure of L-Band EDFA with two-stage pumps, which utilizes the forward ASE as a second pump.

本论文本着科技创新、服务于生产实践的精神，以设计价格低廉，性能优越、结构合理简单的L-band EDFA为目标，依托浙江省科技厅重大科学基金资助项目(001101027)和浙江大学振兴教育计划资助的全光网络的搭建—《光通信技术》课程实验建设项目，展开了如下工作：首先，介绍了掺铒光纤放大器的工作原理，归纳总结了基于速率方程基础的EDFA的各种理论模型，分析比较了各个模型的特点和适用场合，最终选择完善的Giles理论模型作为本论文分析设计EDFA的理论基础；针对实际实验条件提出了修正的EDFA模型的数值计算方法；创新地提出了增益—噪声系数全局分析法，直观有效地分析了EDFA的增益和噪声系数与掺铒光纤长度和泵浦功率的依赖关系，并对各种EDFA的性能作了全面的比较；第二，介绍了掺铒光纤放大器特性参数的定义与测量方法，重点介绍了噪声系数和增益系数的测量方法及实验装置；介绍了组成掺铒光纤放大器各组件的特性测量方法，重点介绍了测量掺铒光纤参数的方法和实验装置；第三，针对WDM系统对L-band EDFA谱平坦、低噪声、高增益的要求，我们创新地提出了四种不同的L-band EDFA结构，按设计思路可以分为两大类：一类是基于单抽运二级泵浦法，这一类包括在未泵浦掺铒光纤的输入端插入一根布拉格光纤光栅的两段L-band的EDFA的新结构；基于前向ASE光作为二级泵浦源推动下一级EDF工作的泵浦分配、两段级联L-band的EDFA的新结构；基于单根光纤光栅、泵浦分配、两段级联的EDFA；另一类则是基于同时应用前后向C-band ASE作二级泵浦源的双抽运法，如基于前后向ASE光作为二级泵浦源的三段级联L-band的EDFA的新结构。

For metal notch strength estimation, either Neuber method based on Neuber's rule or energy method based on Moski and Glinka's equivalent energy density method may be used.

估算金属切口的强度时，人们一般采用以Neuber准则为理论基础的Neuber法，或以Moski及Glinka的等效能量密度法为理论基础的能量法。

Do you know, i need you to come back

你知道吗，我需要你回来

Yang yinshu、Wang xiangsheng、Li decang,The first discovery of haemaphysalis conicinna.

1〕 杨银书，王祥生，李德昌。安徽省首次发现嗜群血蜱。