Title :
Research on Fault Diagnosis Based on Wavelet Packet Multi-class Classification SVM
Author :
Xiaogang Xu ; Songling Wang ; Fei Li ; Zhengren Wu ; Wei Sun
Author_Institution :
Dept. of Power Eng., North China Electr. Power Univ., Baoding, China
Abstract :
It\´s still in search that how to apply SVM in muti-classify. Directed Acyclic Graph is easier to be computed and has better learning effect than other arithmetic. Experimental platform is used to simulate typical faults of circumrotate machines. Based on the frequency domain feature, energy eigenvector of frequency domain is presented using wavelet packet analysis method. DAGSVM is applied in classification, a "grid-search" is applied on C and using cross-validation. The classify effect is more veracious that of BP network.
Keywords :
directed graphs; fault diagnosis; learning (artificial intelligence); pattern classification; search problems; support vector machines; SVM; circumrotate machine; cross validation method; directed acyclic graph; energy eigenvector; fault diagnosis; fault simulation; frequency domain feature; grid search; learning effect; multiclass classification; wavelet packet analysis; DAGSVM; fault diagnosis; pattern matching; rotating machine; wavelet packet;
Conference_Titel :
Manufacturing Automation (ICMA), 2010 International Conference on
Conference_Location :
Hong Kong
Print_ISBN :
978-1-4244-9018-9
Electronic_ISBN :
978-0-7695-4293-5
DOI :
10.1109/ICMA.2010.41
