Table 9 Benchmark comparison of fault diagnosis methods for mechanical vibration Signals.
From: Real time fault diagnosis in industrial robotics using discrete and slantlet wavelet transformations
Method | Application | Accuracy (%) | Processing Time (sec.) | Validation Method | Reference |
|---|---|---|---|---|---|
CNN-based (Li et al., 2019) | Bearing fault diagnosis | 99.80 | 11.2 | 10-fold CV | |
LSTM-based (Zhao et al., 2019) | Rotating machinery | 99.40 | 12.5 | 5-fold CV | |
A Hybrid Adaptive Fusion DL Model (JUNYU REN et al., 2025) | Bearing fault detection | 89.81 | - | 5-fold CV | |
Transformer-based (Li et al. ,2024) | Composite fault diagnosis of rolling machinery | 94–96 | - | a separate validation | |
EGN-OOD (Li et al., 2024) | Intelligent machinery OOD fault | 100.00 | 9.5 | 5-fold CV | |
SLT-MLP-ANN (This study) | Robotic arm multi-joint faults | 100.00 | 3.73 | 5-fold CV | This work |
DWT-MLP-ANN (This study) | Robotic arm multi-joint faults | 100.00 | 7.83 | 5-fold CV | This work |
