一种基于小波包和PCA的超声缺陷识别方法
|
摘要
在超声检测中,对缺陷进行定性分析是无损检测与评价的关键内容。本研究提出一种对缺陷类型进行分类的检测方法,通过对不同类型的缺陷波信号进行特征量提取,实现对缺陷的类型识别。首先使用空气耦合超声检测系统采集无缺陷信号与3种不同类型的缺陷波信号,提取信号的时域无量纲参数和小波包能量系数组成多维特征向量;然后使用主成分分析法(Principal component analysis,PCA)对多维特征向量进行降维处理得到特征融合量;最后输入BP神经网络系统进行缺陷类型的分类,并与未经过PCA处理的测试结果进行对比分析。实验结果证明,经过PCA处理的测试结果准确率更高,测试时间更短。
In ultrasonic testing, qualitative analysis of defects is the key content of nondestructive testing and evaluation. In this paper, a detection method of defect classification is proposed, in which the recognition of the defect type is realized by extracting characteristic quantities of different defect wave signals. Firstly, the air-coupled ultrasonic detection system is used to collect defectfree signals and three different types of defect-wave signals, and the time-domain dimensionless parameters and wavelet packet energy coefficients of the signals are extracted to form multi-dimensional feature vectors. Then principal component analysis(PCA)is used to reduce the dimensionality of multi-dimensional feature vectors to obtain the feature fusion quantity. Finally, BP neural network system is input to classify the defect types, and compare with the test results without PCA processing. The experimental results show that the PCA treatment has higher accuracy and shorter test time.
In ultrasonic testing, qualitative analysis of defects is the key content of nondestructive testing and evaluation. In this paper, a detection method of defect classification is proposed, in which the recognition of the defect type is realized by extracting characteristic quantities of different defect wave signals. Firstly, the air-coupled ultrasonic detection system is used to collect defectfree signals and three different types of defect-wave signals, and the time-domain dimensionless parameters and wavelet packet energy coefficients of the signals are extracted to form multi-dimensional feature vectors. Then principal component analysis(PCA)is used to reduce the dimensionality of multi-dimensional feature vectors to obtain the feature fusion quantity. Finally, BP neural network system is input to classify the defect types, and compare with the test results without PCA processing. The experimental results show that the PCA treatment has higher accuracy and shorter test time.
引文
[1]袁鑫超.碳纤维复合材料多层内部结构及缺陷检测方法研究[D].成都:电子科技大学,2018:23-24.
[2]葛亮,苗瑞,葛秋原,等.窄搭接焊缝涡流信号的特征提取与缺陷识别[J].中国机械工程,2019,30(2):225-229.
[3]李茂,杨录,张艳花.基于EMD及主成分分析的缺陷超声信号特征提取研究[J].中国测试,2018,44(2):118-121.
[4]戴波,赵晶,周炎.超声波管道内检测腐蚀缺陷分类识别研究[J].机床与液压,2008,36(7):194-198.
[5]郑善朴,陆铭慧,王俊东,等.多层结构中脱粘缺陷的超声检测方法[J].应用声学,2019,38(1):134-141.
[6]李娟,杨录,张艳花.基于EMD与欧氏距离的缺陷信号识别方法[J].山西电子技术,2011(2):35-36.
[7]杨琳瑜,于润桥,卢超,等.基于BP神经网络的复合材料超声波检测缺陷类型识别[J].无损检测,2007,29(8):450-452.
[8]李小娟.TFT-LCD面板缺陷分类方法研究[D].合肥:合肥工业大学,2015:34-37.
[9]马钢,李俊飞,白瑞,等.基于PCA-SVR模型的油气管道剩余强度预测[J/OL].油气储运,2018-12-13.http://kns.cnki.net/kcms/detail/13.1093.TE.20181211.1631.010.html
[10]唐东林,魏子兵,潘峰,等.基于PCA和SVM的管道腐蚀超声内检测[J].传感技术学报,2018,31(7):1040-1045.
[11]张晓风.基于主成分分析的软件缺陷预测研究[D].南京:南京航空航天大学,2017:13-15.
[12]胡文刚,刚铁.基于超声信号和图像融合的焊缝缺陷识别[J].焊接学报,2013,34(4):53-56.
[13]孙洁娣,靳世久.基于小波包能量及高阶谱的特征提取方法[J].天津大学学报,2010,43(6):562-566.
[14]许桢英,罗来齐,王匀,等.基于PCA的管道缺陷导波信号特征优化方法[J].电子测量技术,2016,39(4):160-163.
[15]卓东,景军锋,张缓缓,等.基于卷积神经网络的短切毡缺陷分类[J].激光与光电子学进展,2019,56(10):1-14.
[2]葛亮,苗瑞,葛秋原,等.窄搭接焊缝涡流信号的特征提取与缺陷识别[J].中国机械工程,2019,30(2):225-229.
[3]李茂,杨录,张艳花.基于EMD及主成分分析的缺陷超声信号特征提取研究[J].中国测试,2018,44(2):118-121.
[4]戴波,赵晶,周炎.超声波管道内检测腐蚀缺陷分类识别研究[J].机床与液压,2008,36(7):194-198.
[5]郑善朴,陆铭慧,王俊东,等.多层结构中脱粘缺陷的超声检测方法[J].应用声学,2019,38(1):134-141.
[6]李娟,杨录,张艳花.基于EMD与欧氏距离的缺陷信号识别方法[J].山西电子技术,2011(2):35-36.
[7]杨琳瑜,于润桥,卢超,等.基于BP神经网络的复合材料超声波检测缺陷类型识别[J].无损检测,2007,29(8):450-452.
[8]李小娟.TFT-LCD面板缺陷分类方法研究[D].合肥:合肥工业大学,2015:34-37.
[9]马钢,李俊飞,白瑞,等.基于PCA-SVR模型的油气管道剩余强度预测[J/OL].油气储运,2018-12-13.http://kns.cnki.net/kcms/detail/13.1093.TE.20181211.1631.010.html
[10]唐东林,魏子兵,潘峰,等.基于PCA和SVM的管道腐蚀超声内检测[J].传感技术学报,2018,31(7):1040-1045.
[11]张晓风.基于主成分分析的软件缺陷预测研究[D].南京:南京航空航天大学,2017:13-15.
[12]胡文刚,刚铁.基于超声信号和图像融合的焊缝缺陷识别[J].焊接学报,2013,34(4):53-56.
[13]孙洁娣,靳世久.基于小波包能量及高阶谱的特征提取方法[J].天津大学学报,2010,43(6):562-566.
[14]许桢英,罗来齐,王匀,等.基于PCA的管道缺陷导波信号特征优化方法[J].电子测量技术,2016,39(4):160-163.
[15]卓东,景军锋,张缓缓,等.基于卷积神经网络的短切毡缺陷分类[J].激光与光电子学进展,2019,56(10):1-14.