摘要
将电磁无损检测原理与聚类分析相结合,利用改进的神经网络聚类学习方法,对钢铁材质进行检测,结果证明,该方法作为一种新的自适应模式识别技术,比传统的电磁检测准确度高,误判率低。
This paper uses a modified cluster analysis learning to resolve the problem of testing on the quality of steel and iron, with the theory of electromagnetic nondestructive testing, integrating the technique of cluster analysis. The experiments show that this method used as an adapting mode of pattern recognition technique has abetter mode classification capability as compared to conventional nondestructive testing method. It is high in accuracy and low in error.
出处
《信息技术》
2004年第5期56-58,共3页
Information Technology
关键词
聚类
神经网络
无损检测
螺栓
cluster
neural network
nondestructive testing
bolt
