Study of Punch Die Condition Discrimination Based on Wavelet Packet and Genetic Neural Network

Authors:
Zhigao Luo;Xiang Wang;Ju Li;Binbin Fan;Xiaodong Guo
Affiliations:
College of Mechanical Engineering, Jiangsu University, Zhenjiang, China 212013;College of Mechanical Engineering, Jiangsu University, Zhenjiang, China 212013;College of Mechanical Engineering, Jiangsu University, Zhenjiang, China 212013;College of Mechanical Engineering, Jiangsu University, Zhenjiang, China 212013;College of Mechanical Engineering, Jiangsu University, Zhenjiang, China 212013
Venue:
ISNN '08 Proceedings of the 5th international symposium on Neural Networks: Advances in Neural Networks, Part II
Year:
2008

Citing 2
Cited 0

Temporal updating scheme for probabilistic neural network with application to satellite cloud classification

IEEE Transactions on Neural Networks
A novel continuous-time neural network for realizing associative memory

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

According to the characteristics of the acoustic emission signal which was induced by punch die when It fails, the characteristic parameters of failure signal is determined. The energy eigenvector of signal failure die is extracted by wavelet packet analysis technology, and the comparison between the energy in different frequency bands and total energy is taken as the characteristic parameters. Then a BP neural network is established in which the time factor is considered based on genetic algorithm. The characteristic parameters are used as input specimen, learning and training the network to complete the pattern recognition of model working state. Experiments show that the method can quickly and reliably discriminate the conditions of the punch die and has strong practicability.