A Novel Method of Fault Diagnosis for Rolling Bearing Based on Dual Tree Complex Wavelet Packet Transform and Improved Multiscale Permutation Entropy

Table 2

Classification results of different classifiers with feature vectors extracted by different methods.

Classifier	Testing accuracies with feature vectors extracted by different methods
	DTCWPT; + IMPE; + LLTSA	DTCWPT; + MPE; + LLTSA	DTCWPT + IMPE	IMPE
ELM	100%	98%	99.25%	93%
SVM	99.75%	97.75%	99.25%	91.5%
ANN	99%	96%	98.25%	90.5%
KNNC	99.5%	97.75%	99%	91.5%
Average of four classifiers	99.56%	97.37%	98.94%	91.63%