巴克豪森噪声应力检测的人工神经网络数据处理方法  

The Artificial Neural Networks Data Processing Method for Stress Testing Based on Barkhausen Noise

作　　者：郝爱云[1] 王文军[1] 杨玉敏[1] 

机构地区：[1]河北工程大学,河北邯郸056021

出　　处：《无损检测》2007年第10期575-577,共3页Nondestructive Testing

摘　　要：介绍了人工神经网络技术的原理、学习与训练方法等,以及如何建立巴克豪森噪声(BN)信号与应力间的关系。分析了实际测量结果,说明该数据处理方法能够在BN应力检测中很好地应用。The principle and algorithm of neural networks, and the way how to build the relationship between barkhausen noise（BN） and stress were introduced. It was found in practical testing and the result analysis that the data processing method could be used well on stress testing based on BN.

关 键 词：巴克豪森噪声信号 径向基函数 人工神经网络 MATLAB软件 

分 类 号：TG115.28[金属学及工艺—物理冶金]