基于RBF神经网络的非线性电子器件的建模方法研究

针对当前电路仿真中存在的复杂非线性电子器件的建模问题,利用RBF神经网络在函数逼近时所具有的速度快、精度高的特点,提出了根据RBF网络学习器件的输入输出特性,将得出的网络结构用Pspice中的电路描述语言进行描述,建立了非线性电子器件模型的方法,并验证其有效性.结果表明该方法可作为一种用于非线性器件建模的通用方法.

A neural network method for modeling nonlinear electron devices based on RBF

Aiming at the problem of modeling complicated nonlinear electron devices in circuit simulation,a method,which makes full use of the capacity of high learning speed and high accuracy of RBF,is proposed.The model can be obtained by describing the architecture of the neural network in Pspice,which is achieved by learning the input and output data of the device.The efficiency of this method is proved by experiments.