Classification of AE Sources of Carbon/Epoxy Composite Based on SVM
-
摘要: 对单丝碳纤维/环氧树脂复合材料的试样以及碳布/环氧树脂试样拉伸，对其拉伸过程中产生的声发射源信号利用声发射波形参数作为不同神经网络的输入对其进行模式识别，发现SVM神经网络比BP神经网络在对声发射源信号进行模式识别时具有更高的预测准确率。在增加SVM网络训练集时，SVM网络的预测准确率有较大的提高。利用声发射波形参数与SVM对碳/环氧树脂复合材料进行模式识别得到较好的效果。Abstract: Carbon/epoxy specimens were made and stretched to fracture. In the process, acoustic emission signals were collected and the AE signal parameters were set as the input parameters of the neural network. It is more exact to use SVM network to recognize the different acoustic emission sources than to use the BP neural network. And also, the accuracy of the SVM increases when the number of the training set increases. It is proved that use AE signal parameters and SVM network can recognize the acoustic emission source pattern well.
-
-
[1] 王春静，代云霏. 碳纤维复合材料在航空领域的应用[J].机电产品开发与创新，2010，23(2):14-15. [2] 史峰，王小川，郁磊，等. Matlab神经网络30个案例分析[M].北京：北京航空航天大学出版社，2010. [3] CHRISTIAN U. GROSSE, MASAYASU O. Acoustic emission testing[M]. Berlin Heidelberg: Springer Publications, 2008. [4] 耿荣生，沈功田，刘时风.基于波形分析的声发射信号处理技术[J].无损检测，2002,24(6):257-261. [5] 李孟源，尚振东，蔡海潮，等.声发射检测及信号处理[M].北京：科学出版社，2010. [6] CORTES C, VAPNIK V. Support vector net-work[J]. Machine Learning,1995,20:273-279. [7] ZHAO Qun, PRNCIPE J C. Support vector machines for SAR automatic target recognition[J]. IEEE Transactions on Aerospace and Electronic Systems,2001,37(2):643-645. [8] HAN Li-qun. Recognition of the part of growth of flue-cured tobacco leaves based on support vector machine[C].//7th World Congress on Intelligent Control and Automation. Chong Qing, China: 2008:3624-3627. [9] 施彦，韩力群，廉小亲. 神经网络设计方法与案例分析[M]. 北京：北京邮电出版社，2009.
计量
- 文章访问数: 2
- HTML全文浏览量: 0
- PDF下载量: 2
下载:
