Author Affiliations
1Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, Yunnan , China2Institute of Intelligent Manufacturing, Guangdong Academy of Sciences, Guangdong Key Laboratory of Modern Control Technology, Guangzhou 510070, Guangdong , Chinashow less
Fig. 1. Process of self-supervised feature extraction
Fig. 2. Power spectrum of pink noise
Fig. 3. Process of pink noise generation by Matlab
Fig. 4. Flowchart of deeply separable convolution
Fig. 5. Structure of Bottleneck module
Fig. 6. Structure diagram of AE
Fig. 7. Flowchart of abnormal sound detection for mechanical equipment
Fig. 8. Time-frequency spectra of test sets of normal sound samples and abnormal sound samples
Fig. 9. Time-frequency spectra of generating abnormal samples
Fig. 10. Visualization of original time-frequency characteristics of test samples
Fig. 11. Visualization of test sample self-supervised feature extraction
Fig. 12. AUC results of six feature extraction methods for anomaly detection
Operator | Expansion factor | Channel | Repeated times | Stride |
---|
Conv2D | - | 16 | 1 | 2 | Bottleneck | 1 | 8 | 1 | 1 | Bottleneck | 6 | 16 | 2 | 2 | Bottleneck | 6 | 16 | 3 | 2 | Bottleneck | 6 | 32 | 4 | 2 | Bottleneck | 6 | 48 | 3 | 1 | Bottleneck | 6 | 80 | 3 | 2 | Bottleneck | 6 | 160 | 1 | 1 | Conv2D | - | 1280 | 1 | 1 | Avg pool | - | 1280 | 1 | - |
|
Table 1. Network structure of MoblienetV2
Type | Number of samples in training sets | Number of normal samples in testing sets | Number of abnormal samples in testing sets |
---|
Slider | 2804 | 400 | 890 | Valve | 3291 | 400 | 479 | Pump | 3349 | 400 | 456 | Fan | 3675 | 400 | 1475 |
|
Table 2. Data partition of four kinds of machine sound
Model | Slider DAUC /% | Valve DAUC /% | Pump DAUC /% | Fan DAUC /% | Average DAUC /% |
---|
SSFE-IF | 94.6 | 85.2 | 82.9 | 74.5 | 84.3 | SSFE-Kmeans | 93.8 | 82.7 | 80.2 | 73.7 | 82.6 | SSFE-OCSVM | 85.6 | 78.8 | 74.5 | 65.3 | 76.1 | SSFE-GMM | 94.5 | 89.9 | 87.6 | 76.8 | 87.2 | SSFE-DAGMM | 94.5 | 91.3 | 88.5 | 78.1 | 88.1 | SSFE-CAE | 94.8 | 91.2 | 89.0 | 78.2 | 88.3 | SSFE-AE | 95.0 | 92.7 | 88.2 | 78.0 | 88.5 |
|
Table 3. Performance comparison of different anomaly detection models