权系数

其次，用最速下降法对PID三个权系数进行调整。

Secondly , to adjust the PID three weight coefficient by using the gradient descent algorithm .

用遗传算法优化BP神经网络的关键有三点：1、染色体位串与权系数的编码映射；

There are three key technologies for the optimization : 1 , chromosome spaces string is shined upon the weight coefficient ;

多目标评价中的一个重要因素是考虑目标权系数的选取。

In this paper , a combination method that allots multi-objective weights is proposed .

用于权系数调节的基于delta规则的算法就是如此。

A delta rule-based algorithm for weight modification looks like this .

第2部分：在输入节点i和输出节点n之间权系数改变，如下所示

Second part : the weight change between input and output nodes , I and n , is

然后，再运用BP算法优化模糊神经网络的连接权系数。

Finally , the back propagation algorithm was used to optimize the connection coefficients of fuzzy neural network .

提出了一种准则权系数和决策者权系数信息都不完全，且决策者存在风险态度的Fuzzy群体多准则决策方法。

A new method of fuzzy group multiple criteria decision making ( FGMCDM ) with incomplete information of criteria 's weights and decision-maker 's is proposed .

引入混合核函数后的SVM多了一个新的可调参数，即权系数。

The introduction of mixed kernel function makes the SVM add an adjustable parameter named weight coefficient .

提出了一种基于熵权系数与vague集的多目标决策方法。

An approach to multiple objectives decision making based on entropy coefficient and vague set is proposed .

delta规则算法总是在权系数上调整，而且这是建立在相对输出的激活方式上。

The delta rule algorithm always makes a change in weights , and it is based on activation as opposed to output .

基于熵权系数与TOPSIS集成评价决策方法的研究

Research on TOPSIS integrated evaluation and decision method based on entropy coefficient

利用GIS技术获得各项评价因子，并进行标准化处理，采用层次分析法和熵权系数法求得各评价指标的综合权重，构建出高标准基本农田评价体系。

Acquire the comprehensive weight of each evaluation index by analytic hierarchy process and entropy Coefficient Method . Structure the evaluation system of high standard basic farmland .

我们利用p级数方法求解多目标规划问题MOP，并用分层法的思想确定权系数。

We use p series method to solve the multi objective programming problem MOP , and put the idea of hierarchy onto the ascertain of function power .

样条微分求积法是一种新型的数值计算方法，其与传统微分求法的主要区别在于：它是基于B样条函数来构造基函数，进而获得权系数。

The Spline-based Differential Quadrature ( SDQ ) is a newly developed numerical method . The main distinction of the SDQ lies in the determination of weighting coefficients on the basis of cardinal B-spline interpolation functions .

该模型利用模糊聚类技术确定系统的模糊空间和模糊规则数，利用BP算法调整模糊神经网络的权系数。

The fuzzy space and the number of fuzzy rules of this model are defined by the fuzzy clustering method and weight coefficients of the model are adjusted by the BP algorithm .

针对德尔菲法测定权系数之不足，提出了对层次分析法（AHP）进行改造并用于确定权系数的构想。

For surmounting the defects of Delphi method , a new design to improve the Analytic Hierarchy Process ( AHP ) and determine the weight coefficient is given .

在考虑3种定位方法精度条件下，给出了传感器的优先权系数计算方法与基于优先权系数的D-S证据融合信度函数；

According to different precision , the sensor priority factor is calculated , based on which the belief function is obtained .

给出了确定n次有理Bézier曲线权因子的权系数极大化方法和幂指数型权因子方法。

A weights maximization method and a method based on the weights of a power exponent form for determining the weights of the rational B é zier curves of degrees are given .

采用LM学习算法对网络进行训练以获得各网络的连接权系数和阈值。

Using the LM learning algorithm to train the network , the connection weights and thresholds of each network was obtained .

采用MATLAB软件作为模型的仿真软件，通过该软件训练网络，从而确定输入数据模糊化的隶属度函数和网络的连接权系数（或模糊关系矩阵）。

The simulation tests of training fuzzy neural network expert system are implemented using MATLAB software . As a result , membership function and synaptic weight coefficient ( fuzzy relationship matrix ) are determined .

利用Lyapunov理论设计了网络权系数阵的在线学习规则，并证明了观测器的稳定性。

Furthermore , the Lyapunov function is created , and on-line learning rules are given for the network weight matrix , such that the stability of observer is proved .

基于R型因子分析，作者在文中提出了定量确定单指标权系数的加权平均意义下的综合指标和因子得分意义下的综合指标，并且指出了在特定条件下这二种指标的归一性。

Two rational combined indices are put forward with a weighted coefficient of single index under both weighted average and factor score graded by type R factor analysis , and these combined indices have the same meaning .

文中叙述了算法的基本原理及仿真结果，指出了对于LFM信号，修正权系数的LMS算法比LMS算法具有更好的提取弱信号的能力，并且算法运算量小。

The basic principle of the proposed algorithm and the simulation results are described . The simulation results show that the modified LMS algorithm is satisfactory .

本文讨论了各种权系数方法，目前主流的赋权方法主要分为两类，一类是主观赋权法，主要是以Delphi法，AHP法，专家会议法为代表。

The current mainstream approach enabling divided into two main categories . One category is subjective weighting methods , based on Delphi method , AHP , and the Expert Meeting method .

制造业；供应商选择；层次分析法；熵权系数法；TOPSIS法；

Manufacturing industry ; Supplier selection ; Analytic Hierarchy Process ; Entropy Coefficients ; Technique for order preference by similarity to ideal solution ( TOPSIS );

针对这一特点，本文面向实际应用需求，提出一种基于熵权系数法的无线传感器网络自适应QoS路由算法（AnEntropy-basedAdaptiveQoSRouting，EAQR）。

Based on this , oriented on practical application requirements , an entropy-based adaptive QoS routing ( EAQR ) algorithm for wireless sensor networks was proposed in this paper .

最后的仿真结果表明，基于SVM的组合预测模型的预测精度不仅高于任一单一模型，且高于固定权系数组合预测模型。

Simulation results indicate that accuracy of the combination forecasting model based on SVM proposed in the paper is higher than any one sole model and higher than the stationary weights combination forecasting model .

TLU通过改变它的权系数和阈值来学习。

A TLU learns by changing its weights and threshold .

推导了信噪比增益与降采样滤波器权系数的关系，给出了采用Rayleigh商方法求解给出准则下最优FIR滤波器的方法和步骤。

The relation of the SNR and the coefficients of the filter adopted is deduced . The steps by the Rayleigh quotient method to obtain the optimal desampling FIR filter are also presented .

通过引入单参数，应用权系数的方法，精确估算了一个权系数的不等式，从而对一个Hilbert型不等式进行了具有最佳常数因子的推广。

By introducing a parameter and using the way of weight coefficient , we computed accurately a weight coefficient and gave a generalized Hilbert-type integral inequality with a best constant factor .