Some numerical examples are carried out by using the efg method . the imposition of essential boundary conditions by lagrange multipliers is imperative . the rate of convergence when changing the data of weight function and how to select the data are shown 编制了相应程序，通过算例表明了拉格朗日乘子对强制边界条件的作用及无网格伽辽金方法在权函数参数变化时的收敛特性，论证了无网格伽辽金-有限元耦合的有效性。
In addition , based on pcc ( principle component classifier ) , we study a kind of kernel - based classifier . adopting the proper weighted functions , we obtain wpcc ( weighted pcc ) which has better generalization and ability to deal with the outliers than pcc 此外基于pcc ( principlecomponentclassifier )研究了一类基于核的分类器，通过在pcc中引入加权函数得到抗野值能力更强和推广性能更优的加权wpcc ( weightedpcc ) 。
3 . a kind of urban traffic coordination control arithmetic based on distributed q - learning is designed for the distributed control structure of multi - agent . and designed a kind of reward function and distributed weight function , which are fit to the fields of traffic control 3 .针对multi - agent框架体系的分布式控制结构,设计了一种分布式q学习算法,并且提出了一种适用于交通控制领域的奖惩函数以及分布权值函数。
By introducing a set of appropriate orthogonal basis functions into the input space , the input functions and the weight functions are expanded under the orthogonal basis functions , and the time aggregation operation of the process neurons is simplified by using the orthogonality of the basis functions 在输入空间中引入一组合适的函数正交基，将输入函数和网络权函数表示为该组正交基的展开形式，并利用基函数的正交性简化网络聚合运算过程。
To analyze some key technologies of optical network relative with rwa in detail , such as transmission , switching and internetworking ; to emphasize on the research of function , fabric and performance of optical cross - connection ; to carry out numerical simulations for crosstalk introduced by optical cross connect and to present measurements for suppressing it such as doubly filtering , fixing optimum decision threshold and appropriately choosing the number of multiplexed wavelengths ; 3 . to research the fundamental principle and some problems relative with rwa , including the type of optical network , the type of traffic , the type of service , the survivability of optical network ; to classify and compare rwa algorithms and particularly research some dynamic rwa algorithms ; 4 . to present reserved light - path and classify network resource such as used , unused and reserved status , to emulate establishment of all - optical connection in optical network through modified rwa algorithm and show effectively reducing setup time of all - optical connection utilizing reserved light - path ; to research rwa algorithms of multi - fiber network , to present new link weight functions dependent on node degree , unused fiber ( s ) per wavelength - layer and routing policies , to perform emulation of rwa based wavelength layer graph applying new link weight functions and show them make algorithms better performance and network lower blocking rate ; 5 详细分析了与rwa相关的光网络关键技术，包括传输、交换、组网等，重点研究了光交叉连接的功能和结构、性能，对其引入的串扰进行了详细分析，选择恰当的器件参数进行了数值模拟，并提出了抑制措施(如双重滤波、优化判决门限、选择恰当的复用波长数) ; 3 .研究了光网络的r认叭的基本原理、与r认叭的几个相关问题(光网络类型、业务类型、流量类型、光网络生存性) 、 r认人算法的分类和比较，具体研究了几种动态r场人算法; 4 .研究了以全光连接建立时间为优化目标的r认认算法，提出预置光路的概念，对网络资源进行状态分类(占用、未占用、预置) ，利用改进的r认叭算法仿真，预置光路可为部分新到的连接请求快速建立连接，从而提高网络性能;研究了以多光纤网络连接阻塞率为优化目标的r认城算法，提出了以节点度数、每个波长分层的空闲光纤数以及路由策略决定的几种链路权重函数，利用基于波长分层图模型的并行r场人算法仿真，利用新的链路权重函数使得算法具有更优的性能，使网络具有更低的连接阻塞率。
It redefines the between - class scatter by adding a weight function according to the between - class distance . therefore , it can separate the classes as much as possible . at the same time , it projects the between - class scatter into the null space of the within - class scatter that contains the most discriminant information 新方法重新定义了样本类间离散度矩阵，在原始定义的基础上增加一个由类间距离决定的可变权函数，使得在选择投影方向时，能够更好地分开各个类的样本。
