Important Note. The second column of the file represents speed. However, since we collected the data at a constant speed, this column is redundant and holds no meaningful value. You can ignore it. The correct speed is indicated in the filename.
We released an open-source bearing failure dataset and hope this dataset can benefit your research
Data of quark disk: https://pan.quark.cn/s/0c7578963014
Data of Google Drive: https://drive.google.com/drive/folders/1UMOvyfstYJRyR0rPw0OfH-tIjJg2_0aN?usp=sharing
重要注释: 文件中的第二列表示转速。然而，由于振动信号是在恒定转速条件下的，该列是冗余的，没有实际意义，可以忽略。实际的转速信息请以文件名为准。
我们发布了一个开源的轴承故障数据集，希望该数据集能够对您的研究有所帮助。
夸克网盘数据下载： https://pan.quark.cn/s/0c7578963014
Google Drive 数据下载： https://drive.google.com/drive/folders/1UMOvyfstYJRyR0rPw0OfH-tIjJg2_0aN?usp=sharing
This dataset comprises vibration signals from bearings in 9 different health states under 11 distinct operating conditions. These datasets are publicly available, and anyone can use them to validate diagnosis algorithms for rolling element bearings. Publications making use of the HUSTbearing datasets are requested to cite the following paper.
Chao Zhao, Enrico Zio, Weiming Shen, Domain Generalization for Cross-Domain Fault Diagnosis: an Application-oriented Perspective and a Benchmark Study, Reliability Engineering and System Safety (2024), doi: https://doi.org/10.1016/j.ress.2024.109964.
该数据集包含在 11 种不同工况下、9 种不同健康状态的轴承振动信号。这些数据集是公开可获取的，任何人都可以使用它们来验证滚动轴承故障诊断算法。凡是使用 HUSTbearing 数据集发表的研究成果，请引用以下论文：
Chao Zhao, Enrico Zio, Weiming Shen, Domain Generalization for Cross-Domain Fault Diagnosis: an Application-oriented Perspective and a Benchmark Study, Reliability Engineering and System Safety (2024), doi: https://doi.org/10.1016/j.ress.2024.109964.
Fig. 1. (a) Test rig of HUSTbearing dataset.
The bearing fault tests were conducted using a Spectra-Quest Mechanical Fault Simulator, as depicted in Fig. 1. From left to right on the test rig are speed control, motor, shaft, acceleration sensor, bearing, and data acquisition board. The bearings in nine health states are illustrated in Fig. 2, representing
(1) normal,
(2) medium inner race fault,
(3) severe inner race fault,
(4) medium outer race fault,
(5) severe outer race fault,
(6) medium ball fault,
(7) severe ball fault,
(8) medium combination fault, and
(9) severe combination fault.
It's important to note that combination fault denotes a fault in both the inner race and outer race. All faults are artificially preset.
| Fault size | severe | medium |
|---|---|---|
| Inner/outer race | 0.3 mm | 0.15 mm |
| Ball | 0.5 mm | 0.25 mm |
Fig. 2. Photographs of the failure bearings.
The type of tested bearings is ER-16K, and the detailed parameters are given in Table 1.
Table 1. Parameters of the tested bearings
| Parameter | Value |
|---|---|
| Shaft Diameter | 38.52 mm |
| Ball Diameter | 7.94 mm |
| Number of Balls | 9 |
A total of 4 different operating conditions were set in experiments. The operating conditions include:
- 20 Hz
- 25 Hz
- 30 Hz
- 35 Hz
- 40 Hz
- 60 Hz
- 65 Hz
- 70 Hz
- 75 Hz
- 80 Hz
- 0-40-0 Hz,
Fig. 3. Time-varying speed. （时变转速示意图）
The sampling frequency is set to 25.6 kHz. A total of 262144 data points (i.e. 10.2s) are recorded for each sampling.
The raw data file comprises 99 files (9 health states multiplied by 11 working conditions), each in Excel format.
For instance, the filename "0.5X_B_65Hz" indicates a medium ball fault under the 65 Hz working condition, where 0.5X denotes medium damage.
The health states are represented by the following codes:
H: healthy
I: inner race fault
O: outer race fault
B: ball fault
C: combination fault
For example, "O_80Hz" signifies a severe outer race fault under the 80 Hz working condition.
Fig. 4. Drive motor information (驱动电机信息).
If you have any questions or suggestions, do not hesitate to contact:
Mr. Chao Zhao, zhaochao0612@gmail.com
