Fault Diagnosis Method for Industrial Robots Based on DBN Joint Information Fusion Technology

Table 3

Performance comparison of different fault diagnosis models.

Diagnostic models	Training set accuracy (%)	Test set accuracy (%)	Running time (s)	Standard deviation
DBN	87.34	81.66	61.63	0.0703
VMD + BP	92.72	86.58	39.17	0.0418
VMD + SVM	—	92.08	20.07	0.0658
EMD + DBN	90.21	86.17	48.55	0.0277
EMD + SVM	—	94.25	17.43	0.0491
The proposed method	99.12	97.96	23.46	0.0074