A Novel Method for Intelligent Fault Diagnosis of Bearing Based on Capsule Neural Network

Table 4

Average classification accuracy of different algorithms.

Average accuracy of test sample set (%)	XCN	DWAE+ELM	CapsNet	MPE+ISVM+BT	AE+ES+CNN	DBN
Inner ring fault	97.2	88.6	95.6	87.6	90.4	89.1
Outer ring fault	98.7	91.3	94.3	90.3	92.6	91.3
Tapered roller fault	94.5	88.6	91.9	92.3	87.2	84.3