薄膜场发射开启电场的小波神经网络预测模型研究  

Wavelet neural networks model on threshold electric field of field emission from thin films

作　　者：崔万照[1] 朱长纯[1] 刘君华[1] 

机构地区：[1]西安交通大学电信学院电子系,西安710049

出　　处：《物理学报》2004年第5期1583-1587,共5页Acta Physica Sinica

摘　　要：基于小波变换的特征信息的提取和神经网络的自学习能力研究了薄膜场发射的特性 ,结合阴极薄膜场发射的特点 ,建立了薄膜场发射开启电场的小波神经网络预测模型 .并用金刚石薄膜场发射开启电场数据进行了验证 ,结果表明该模型预测的相对误差小于 1 30 % .这一结论预示着小波神经网络是一种研究薄膜场发射特性的方法 .In this paper, characteristics of field emission from thin films have been studied based on extracting features of wavelet transforms from experimental data and neural networks self-learning,and the forecasting model of wavelet neural networks in combination with the characteristics of cold cathode materials has been established. The data of diamond thin film on threshold electric field in field emission is used to test this model, and the results show that the absolute value of the relative error of the model is within 1.30%. This result implies that the wavelet neural network is a useful tool for studying characteristics of field emission from thin films.

关 键 词：小波变换 神经网络 开启电场 场发射 场发射显示器 薄膜 

分 类 号：O484.4[理学—固体物理]