File Names
The filenames for the vibration data follow the format vibration_[mode]_[num1]_[num2].csv, where:
[mode] describes the condition of the bearing and can be normal, inner, outer, or ball.[num1] and [num2] are arbitrary numbers with no specific meaning.Added Noise
To create a dataset similar to real-world industrial environments, Weibull Noise and background noise from the ToyADMOS dataset have been added to the vibration signals at various signal-to-noise ratios.
Weibull Noise consists of signals with a signal-to-noise ratio (SNR) of -50 dB, shape factors between 1.5 and 2, and scale factors between -2 and 2.
Additionally, ToyADMOS Noise has an SNR ranging from 0 dB to -4 dB, with noise randomly selected from over 380 different types of signals added to both the x-axis and y-axis data.
Data Specification (Train + Evaluation Dataset)
| Data Types | Speed Condition | Fault Location | Fault Types | Number of Files | Load (Nm) | Rotating Speed (RPM) | Sampling Rate (kHz) |
|---|---|---|---|---|---|---|---|
| Vibration | Varying | Bearing Housing B | Normal | 1,470 (train) + 630 (eval) | 0 | 680 ~ 2460 | 25.6 |
| Vibration | Varying | Bearing Housing B | Inner | 630 (eval) | 0 | 680 ~ 2460 | 25.6 |
| Vibration | Varying | Bearing Housing B | Outer | 630 (eval) | 0 | 680 ~ 2460 | 25.6 |
| Vibration | Varying | Bearing Housing B | Ball | 630 (eval) | 0 | 680 ~ 2460 | 25.6 |
(Test Dataset): Same conditions as the train dataset, total 1200 files, with the ratio of normal to faulty data undisclosed