Vibration based Fault Diagnosis of Machines

This dataset contains vibrational data collected for the purpose of diagnosing early faults in machinery. The data was used in the research work titled "System Design for Early Fault Diagnosis of Machines using Vibration Features" published at 2019 International Conference on Power Generation Systems and Renewable Energy Technologies (PGSRET). In order to use this dataset, please cite this article:  M. U. Khan, M. A. Imtiaz, S. Aziz, Z. Kareem, A. Waseem and M. A. Akram, "System Design for Early Fault Diagnosis of Machines using Vibration Features," 2019 International Conference on Power Generation Systems and Renewable Energy Technologies (PGSRET), Istanbul, Turkey, 2019, pp. 1-6, doi: 10.1109/PGSRET.2019.8882726.

Data Collection:

Sensor Used: SG-Link tri-axial accelerometer sensor by MICROSTRAIN Corporation.

Sampling Rate: 679 samples per second for each of the three axes (axial, horizontal, and vertical).

Experiment Location: Mechanical Vibration Laboratory, Mechanical Engineering Department (MED), University of Engineering and Technology (UET), Taxila.

Working States: The experiment focused on four working states of the machinery:

Normal state

Cracking state

Offset pulley state

Wear state

RPM Values: The data was collected at five different rotational speeds (RPMs):

1262 RPM

1312 RPM

1374 RPM

1410 RPM

1424 RPM

Experimental Setup: The experimental setup involves a test rig apparatus designed to simulate different fault conditions in machinery. The block diagram of the setup is shown in Figure (3) of the associated research paper.Dataset Structure: The dataset is organized into folders corresponding to the three axes of vibration data:

x_axis/ - Contains vibration data along the x-axis (horizontal direction).

y_axis/ - Contains vibration data along the y-axis (vertical direction).

z_axis/ - Contains vibration data along the z-axis (depth or axial direction).

Usage Notes:

The dataset is provided as-is for research and educational purposes.

Please cite the associated paper when using this dataset: https://doi.org/10.1109/PGSRET.2019.8882726