Journal: Data in Brief

Article Title: Multi-domain vibration dataset with various bearing types under compound machine fault scenarios

doi: 10.1016/j.dib.2024.110940

Figure Lengend Snippet:

Article Snippet: Data collection , Raw vibration data were gathered using a USB digital accelerometer (PCB Piezotronics 333D01) from the AST fault simulator at sampling rates of 8 kHz and 16 kHz. Looseness, unbalance, and misalignment were simulated for faults of rotating components. Three types of bearing were used to generate multi-domain datasets: a deep groove ball bearing (6204), cylindrical roller bearings (N204 and NJ204), and a tapered roller bearing (30204), each with the ball, inner-raceway, and outer-raceway faults. Additionally, data were collected at six different rotating speeds. The raw vibration signals were also converted to 2-D spectrograms using the short-time Fourier transform (STFT) by the MATLAB program..

Techniques: Biomarker Discovery, Sampling