Research Article

Discrete Cosine Transformation and Temporal Adjacent Convolutional Neural Network-Based Remaining Useful Life Estimation of Bearings

Table 3

Performance evaluation with different frame numbers.

ā€‰DCT-TACNN/TFR-TACNN

MAE90.85/121.6866.03/100.6645.23/80.3553.31/72.1263.69/79.35
MSE90.45/105.9341.79/80.5936.27/75.0233.15/70.3350.10/88.54
MAPE0.3754/0.52140.3266/0.50250.1893/0.46210.2619/0.42440.3028/0.4459