摘要
提出了一种基于振动的蚕茧质量智能无损检测的方法.通过对蚕茧振动信号进行小波分解重构蚕蛹的振动冲击信号,提取与蚕蛹质量有关的特征值,并利用模糊聚类的方法优选特征值,然后将优选的特征值作为BP神经网络的输入参数,以蚕蛹质量作为输出参数训练该网络,训练后的神经网络可利用所采集的蚕茧振动信号确定蚕蛹的质量,从而间接得到茧壳的质量.试验结果表明该方法有效可行.图4,表2,参9.
A kind of non-destructive test method for silk cocoon based on vibration signal was authors applied wavelet theory to reconstructing chrysalis vibration signals. The features among ch proposed. The rysalis vibration signal,which were related to chrysalis weights,were extracted respectively. And the characteristic parameters were selected firstly by fuzzy classification. Then BP model of neural networks was built,which inputs include seven relevant characteristic parameters. The output of neural network represents weights of the chrysalis. The weight of cocoon shells can be gain indirectly. The experiment results show that the method are effective and feasible. 4figs., 2tabs., 9refs.
出处
《湖南科技大学学报（自然科学版）》
CAS
北大核心
2006年第1期17-20,共4页
Journal of Hunan University of Science And Technology：Natural Science Edition
基金
湖北省"十五"重点科技攻关项目资助(编号:2001AA208B02)
国家"十五"科技攻关项目资助(2001BA502B01-03-02)
关键词
蚕茧
无损检测
振动
神经网络
小波包
特征提取
pod
non-destructive testing
vibration
neural networks
wavelet packet
feature extraction
