摘要
超声检测中对缺陷进行类型分析通常取决于操作人员对于特定专业知识的了解及检测经验,从而导致其分析结果的不稳定性和个体差异性。本文提出了一种使用小波包变换提取缺陷特征信息,并应用深度神经网络对得到的信息进行分类识别的方法。利用超声相控阵系统对于不锈钢试块上的通孔、斜通孔和平底孔进行超声检测,并对得到的超声回波波形按照新方法进行分析。实验结果表明,使用小波包变换后的数据进行分类识别能够在提高识别准确率的同时降低神经网络的学习时间,而使用深度神经网络相比通用的BP神经网络以可接受延长学习时间的代价提高了识别的准确率。采用新方法后,缺陷分类正确率提高了21.66%,而网络学习时间只延长了91.9s。在超声检测中使用小波包变换和深度神经网络来对于缺陷进行类型分析,能够排除人为干扰,增加识别准确率,对于实际应用有着极大的意义。
Defect classification analysis in ultrasonic detection usually depends on the operator's professional knowledge and experience,which leads to the instability and individual differences of the analysis result.To solve the problem,a method for classification analysis of ultrasonic detection signals is presented.The method uses wavelet packet transform to extract the feature information from ultrasonic defection,and applies deep neural networks to classification.Experiments are tested by ultrasonic phased array system to collect detection signal which comes from via holes,inclined holes or flat-bottom holes.Then,classification analysis is done by the new method.Experimental results show two points.The first point is that using wavelet packet transform can increase the recognition accuracy and decrease the learning time of neural networks.The second point is that using deep neural networks can increase the recognition accuracy while increase acceptable learning time comparing to the common BP neural networks.By using the new method,the defect classification accuracy increases up 21.66%while the learning time increases only 91.9 s.Applying wavelet packet transform and deep neural networks to classification analysis in ultrasonic detection can exclude human interference and improve the recognition accuracy.As so,the new method has a great future in practical applications of ultrasonic detection.
引文
1彭应秋.超声波检测工艺.北京:航空工业出版社,1990:5-7
2吴淼,孙海燕,孙智,刘旭.超声检测缺陷分类的小波分析与神经网络方法.中国矿业大学学报,2000;29(3):239-243
3孙芳.超声相控阵技术若干关键问题的研究.天津:天津大学,2012
4 Vong C M,Wong P K.Engine ignition signal diagnosis with wavelet packet transform and multi-class least squares support vector machines.Expert Systems with Applications,2011;38(7):8563-8570
5 Evagorou D,Kyprianou A,Lewin P L et al.Feature extraction of partial discharge signals using the wavelet packet transform and classification with a probabilistic neural network.IET Science,Measurement&Technology,2010;4(3):177-192
6 Bahoura M,Simard Y.Blue whale calls classification using short-time Fourier and wavelet packet transforms and artificial neural network.Digital Signal Processing,2010;20(4):1256-1263
7 Keskes H,Braham A,Lachiri Z.Broken rotor bar diagnosis in induction machines through stationary wavelet packet transform and multiclass wavelet SVM.Electric Power Systems Research,2013;97:151-157
8 Vong C,Wong P,Ip W.Case-based expert system using wavelet packet transform and kernel-based feature manipulation for engine ignition system diagnosis.Engineering Applications of Artificial Intelligence,2011;24(7):1281-1294
9李弼程,罗建书.小波分析及其应用.北京:电子工业出版社,35-38,77-88
10曾宪伟,赵卫明,盛菊琴.小波包分解树结点与信号子空间频带的对应关系及其应用.地震学报,2008;30(1):90-96+106
11王云昌,江波.超声波检测中对缺陷的定性分析.国外金属热处理,2004;25(6):40—41
12陈明.Matlab神经网络原理与实例精解.北京:清华大学出版社,2013:2
13 Rumelhart D E,Hinton G E,Williams R J.Learning representations by back-propagating errors.Nature,1986;323(6088):533—536
14简晓明,李明轩.超声检测中人工神经网络对缺陷定量评价.声学学报,2000;25(1):71—77
15张建生,李明轩.多层粘接结构中脱粘界面的人工神经网络余弦变换谱特征识别.声学学报,2001;26(4):349—354
16宋卫华,王小民,李明轩.电磁超声界面回波瞬时图谱分析与神经网络识别.声学学报,2007;32(4):333—337
17 Bengio Y.Learning deep architectures for AI.Foundations and trends~(?)in Machine Learning,2009;2(1):1—127
18 Hinton G E,Osindero S,Teh Y W.A fast learning algorithm for deep belief nets.Neural Computation,2006;18(7):1527—1554
19 Yu D,Siniscalchi S M,Deng L,Lee C H.Boosting attribute and phone estimation accuracies with deep neural networks for detection-based speech recognition.IEEE International Conference on Acoustics,Speech and Signal Processing,2012:4169—4172
20 Dahl G E,Yu D,Deng L,Acero A.Context-dependent pretrained deep neural networks for large-vocabulary speech recognition.IEEE Transactions on Audio Speech and Language Processing,2012;20(1):30—42
21 Stuhlsatz A,Lippel J,Zielke T.Feature extraction with deep neural networks by a generalized discriminant analysis.IEEE Transactions on Neural Networks and Learning Systems,2012;23(4):596—608
22 Deng L,Hinton G,Kingsbury B.New types of deep neural network learning for speech recognition and related applications:an overview.IEEE International Conference on Acoustics,Speech and Signal Processing,2013:8599—8603
23 Yu D,Deng L.Deep learning and its applications to signal and information processing.IEEE Signal Processing Magazine,2011;28(1):145—149+154
24 Arel I,Rose D C,Kamowski T P.Deep machine learning-a new frontier in artificial intelligence research[research frontier].IEEE Computational Intelligence Magazine,2010;5(4):13—18
25 Hinton G,Deng L,Yu D et al.Deep neural networks for acoustic modeling in speech recognition:The shared views of four research group.IEEE Signal Processing Magazine,2012;29(6):82—97
26 Zhao X Y,Gang T,Zhang B X.Calculation of radiation acoustical fields from phased arrays with nonparaxial multi-Gaussian beam model.Chinese Journal of Acoustics,2009;28(2):146—153
27 Zhang B X,Zhang C G,Deng F Q.Non-integral dimensional ultrasonic phased arrays in a borehole.Chinese Journal of Acoustics,2009;28(3):220—230
28 Shi F F,Wu X M,Zhang B X.Application of cylindrical linear phased array in casing borehole.Chinese Journal of Acoustics,2010;29(1):65—72
29 Li Q F,Shi L H,Liang D K.Research on 2D imaging technique for concrete cross section.Chinese Journal of Acoustics,2010;29(1):85—96
30 Kang T,Kim H J,Song S J.Ultrasonic guided wave inspection with array sensors.Chinese Journal of Acoustics,2010;29(4):377—385