网络函数

A sufficient and necessary condition for nonlinear continuous time associate memory neural networks to be locally exponentially stable is obtained, and a function is defined for characterizing the domain of attraction of equilibrium point.

讨论非线性连续联想记忆神经网络平衡点局部指数稳定的判定条件及平衡点指数吸引域的估计，得到了平衡点局部指数稳定的充要条件，并引入一个特征函数，可以判定平衡点的邻域是否为指数吸引域。

A sufficient and necessary condition for nonlinear continuous-time associate memory neural networks to be locally exponentially stable is obtained, and a unction is defined for characterizing the domain of attraction of equilibrium point.

摘要讨论非线性连续联想记忆神经网络平衡点局部指数稳定的判定条件及平衡点指数吸引域的估计，得到了平衡点局部指数稳定的充要条件，并引入一个特征函数，可以判定平衡点的邻域是否为指数吸引域。

The simulation model of current and speed double loops PMSM control system is made up in Matlab/Simulink.

然后设计了神经网络PID控制器，将电机等效为一个传递函数，使用MATLAB仿真软件中的m文件编程方式进行系统仿真。

Key words:magnetic resonance electrical impedance tomography;magnetic flux density measurement; radius basic function;neural network;genetic algorithm

作者单位：浙江大学电气工程学院，浙江杭州 310027 核磁共振电阻抗成像；磁感应强度测量；径向基函数；神经网络；遗传算法

Objective To develop a new Two-step magnetic resonance electrical impedance tomography algorithm based on radial basic function neural network for imaging electrical impedance distribution of a head.

目的基于径向基函数（radial basic function，RBF）神经网络的两步核磁共振电阻抗成像（magnetic resonance electrical impedance tomography，MREIT）算法，对人体头部进行MREIT。

YAN Dandan,ZHANG Xiaotong,ZHU Shanan,Bin HeAbstract: Objective To develop a new Twostep magnetic resonance electrical impedance tomography algorithm based on radial basic function neural network for imaging electrical impedance distribution of a head.

目的 基于径向基函数(radial basic function，RBF)神经网络的两步核磁共振电阻抗成像(magnetic resonance electrical impedance tomography，MREIT)算法，对人体头部进行MREIT。

During the training of the neural network, the influence, which is caused by the various network parameters, on the error function is discussed emphatically and the traditional BP algorithm is improved by importing the weight coefficient βand error function coefficient α.

在神经网络的训练过程中，在快速BP算法的基础上做了一些有效的改进，如：引入权值系数β，选取相应的误差函数系数α等。

In order to improve the convergence rate of network, many scholars have conducted thorough researches into the structure and application of error function.

为了提高网络收敛速度，许多学者对误差函数的构造与应用进行了深入研究。N.B。

According to the Lyapunov stability theory, the weight update law of neural network, the adaptive law of error bound, the stabilizing function and the control are obtained, thus the stability of closed-loop control system is ensured.

基于Lyapunov稳定性理论，得到了神经网络的权值更新律、误差界自适应律、稳定函数及控制律，确保了闭环控制系统的稳定性。

This is because the effect of component in welding material is very complex and no explicit function expression can be gotten before build up the relationship between the recipes of welding material and its performance.

由于焊接材料配方的组份与性能间作用关系的复杂性，无法事先知道组份与配方性能之间关系的明确分析函数形式，本文选用了神经网络建模方法作为建立焊接材料的配方和性能之间的定量关系的工具。

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。