Network Construction for Bearing Fault Diagnosis Based on Double Attention Mechanism

Table 7

Diagnostic results of each model.

Method	900 rpm (%)	1500 rpm (%)	Average accuracy (%)
SSA-SVM	94.66	96.68	95.67
PSO-RF	98.66	96	97.33
AVSSA-KELM	98	98	98
Attention convolutional	98.67	97.35	98.01
Attention LSTM	99	98.67	98.84
Double attention lacks EMD	99.22	99.33	99.27
1D-CNN-LSTM	99.33	99.33	99.33
Paper method	100	100	100