DUDU-BLDC: Data set for diagnostic of Brushless DC motors with degrading magnets

DUDU-BLDC: Dataset for Diagnostics of Brushless DC motors with Degrading Magnets The DUDU-BLDC dataset provides high-resolution measurements for diagnostic research on brushless DC (BLDC) motors under four operating conditions: healthy, mechanically imbalanced, electrically degraded (partial rotor demagnetization), and combined mechanical + electrical faults. Data were collected from Fein EC48 BLDC motors powered by the original 18 V, 3 Ah Li-ion pack and controlled via the Fein ASCM 18 QM PWM drive (16 kHz). Two synchronized sensor channels—an Allegro ACS712-20 Hall-effect current sensor and a Lasergage A2108/LSR laser tachometer—were sampled at 50 kHz and segmented into 0.8 s windows. Following the TIER 4.0 protocol, the archive includes five CSV files: healthy.csv: feature vectors from nominally healthy motors healthy_zip.csv: mechanical-damage (zip-tie imbalance) cases faulty.csv: electrical-degradation cases faulty_zip.csv: combined mechanical + electrical faults motors.csv: merged file with an added “Class” label (Healthy, Mech_Damage, Elec_Damage, Mech_Elec_Damage) Each record comprises 28 time- and frequency-domain features—mean, std, max, RMS, peak-to-peak, skewness, kurtosis, crest factor, spectral center, spectrum area, and amplitudes at 1×, 2×, and 3× rotational harmonics—for both current (A) and speed (RPM). The dataset is released under a CC BY 4.0 license and includes comprehensive metadata and README instructions for immediate reproducibility and benchmarking of fault-detection algorithms. The more complete description of the dataset can be found in: Jarzyna, K., Piątek, P., Bauer, W., & Baranowski, J., “DUDU-BLDC: Data set for diagnostic of Brushless DC motors with degrading magnets,” accepted at IEEE 7th International Conference on Electrical, Control and Instrumentation Engineering (ICECIE) – Pattaya, Thailand | 22-23 Nov 2025