Experimental Research on the Crack Identification of Drawing Parts Based on AE Technique

Abstract:

Article Preview

The paper performs an experimental research on the crack identification of drawing parts using AE technique. Under the platform of the AE system, the AE signals of drawing parts crack are acquired. BP neural network is designed with three layers. They are ten neurons of input layer, three neurons of output layer and thirteen neurons of hidden layer. The characteristic parameters of the crack acoustic emission are considered as the input of BP neural network to exercise the network. The test data are inputted to the neural network after it is exercised. The test result is in accord with the experiment result. The method is proper to identify the crack of drawing parts. The emergence of many inferior parts and the waste of resource can be avoided. It also can debase the cost of manufacture and improve the productive efficiency.

Info:

Periodical:

Edited by:

Hun Guo, Taiyong Wang, Zeyu Weng, Weidong Jin, Shaoze Yan, Xuda Qin, Guofeng Wang, Qingjian Liu and Zijing Wang

Pages:

275-278

DOI:

10.4028/www.scientific.net/AMM.141.275

Citation:

Z. G. Luo et al., "Experimental Research on the Crack Identification of Drawing Parts Based on AE Technique", Applied Mechanics and Materials, Vol. 141, pp. 275-278, 2012

Online since:

November 2011

Export:

Price:

$35.00

In order to see related information, you need to Login.

In order to see related information, you need to Login.