基于神经网络的激光超声探伤表面波的分类 |
| |
作者单位: | ;1.中北大学信息与通信工程学院;2.中北大学理学院 |
| |
摘 要: | 利用3种神经网络即自组织竞争神经网络、学习向量量化神经网络和概率神经网络对激光超声探伤缺陷表面波进行分类.讨论了3种网络在不同输入情况下的分类效果.实验结果表明,这3种神经网络都可以取得良好的分类效果.
|
关 键 词: | 自组织竞争神经网络 学习向量量化(LVQ)神经网络 概率神经网络(PNN) 缺陷检测 |
Classification of Surface Wave by Laser Ultrasonic Flaw Detection Based on Neural Network |
| |
Institution: | ,School of Information and Communication Engineering,North University of China,School of Science,North University of China |
| |
Abstract: | The surface wave was classified by using three kinds of neural networks,i.e.self-organizing competition neural network,learning vector quantization(LVQ)neural network and probabilistic neural network(PNN).Several experiments on different input situations for the three kinds of neural networks were discussed.Experimental results indicated that three kinds of neural networks had good performances in the classification. |
| |
Keywords: | self-organizing competition neural network LVQ neural network PNN defect detection |
本文献已被 CNKI 万方数据 等数据库收录! |