Table 7 Performance comparison of different methods under varying signal-to-noise ratios.
From: RS-SCBiGRU: a noise-robust neural network for high-speed motor fault diagnosis with limited samples
Methods | 0 | 2 | 4 | 6 | 8 | |||||
|---|---|---|---|---|---|---|---|---|---|---|
Accuracy (%) | F1 (%) | Accuracy (%) | F1 (%) | Accuracy (%) | F1 (%) | Accuracy (%) | F1 (%) | Accuracy (%) | F1 (%) | |
RS-SCBiGRU | 51.41 | 45.41 | 64.48 | 59.68 | 74.10 | 72.67 | 81.45 | 79.55 | 86.51 | 85.66 |
SCBiGRU | 41.94 | 33.61 | 55.20 | 48.53 | 68.81 | 66.89 | 80.45 | 78.30 | 85.97 | 84.28 |
BiGRU | 42.46 | 36.02 | 52.67 | 47.16 | 62.87 | 63.87 | 69.21 | 67.50 | 74.12 | 73.23 |
BiLSTM | 35.16 | 27.78 | 44.08 | 37.79 | 56.41 | 53.30 | 63.37 | 59.97 | 69.67 | 67.34 |
DAMN | 12.63 | 3.09 | 14.08 | 4.01 | 18.55 | 7.53 | 23.01 | 13.01 | 31.23 | 22.68 |
MSCNN | 19.35 | 9.25 | 21.86 | 11.46 | 27.60 | 14.57 | 29.19 | 18.28 | 35.57 | 25.38 |
MSCNN-LSTM | 14.13 | 7.38 | 16.48 | 9.84 | 20.04 | 12.24 | 21.78 | 15.70 | 28.02 | 22.67 |
RNN-WDCNN | 50.25 | 44.48 | 59.28 | 55.40 | 67.97 | 67.57 | 74.56 | 73.23 | 79.21 | 78.32 |
WDCNN | 39.55 | 34.50 | 50.53 | 46.90 | 58.96 | 57.96 | 65.99 | 64.08 | 70.01 | 68.36 |
