Research Article

Bearing Remaining Useful Life Prediction Based on AdCNN and CWGAN under Few Samples

Table 6

MAE and RMSE of different percentages of generated data training model prediction results.

MethodsPercentage of generated data
10%20%40%60%80%90%

SVRMAE0.1420.1560.2320.2750.2470.223
RMSE0.1630.1740.2510.2430.2690.251
DNNMAE0.0950.1020.0970.1130.1580.212
RMSE0.1170.1210.1090.1360.1730.229
LSTMMAE0.0670.0650.0710.0920.1050.127
RMSE0.0850.0820.0940.1040.1220.143
AdCNNMAE0.0380.0290.0390.0740.0850.092
RMSE0.0510.0460.0580.0920.0970.105