Research on Fault Diagnosis Based on Wavelet Packet Multi-class Classification SVM

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.

[1]  Chih-Jen Lin,et al.  A comparison of methods for multiclass support vector machines , 2002, IEEE Trans. Neural Networks.

[2]  Gérard Dreyfus,et al.  Single-layer learning revisited: a stepwise procedure for building and training a neural network , 1989, NATO Neurocomputing.

[3]  Girish Kumar Singh,et al.  Induction machine drive condition monitoring and diagnostic research—a survey , 2003 .

[4]  Shyh-Jier Huang,et al.  Application of Morlet wavelets to supervise power system disturbances , 1999 .

[5]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[6]  Isabelle Guyon,et al.  Comparison of classifier methods: a case study in handwritten digit recognition , 1994, Proceedings of the 12th IAPR International Conference on Pattern Recognition, Vol. 3 - Conference C: Signal Processing (Cat. No.94CH3440-5).

[7]  Nello Cristianini,et al.  Large Margin DAGs for Multiclass Classification , 1999, NIPS.

[8]  Ronald R. Coifman,et al.  Entropy-based algorithms for best basis selection , 1992, IEEE Trans. Inf. Theory.