Sung, Yi-CheYi-CheSungYi, Cheng-PeiCheng-PeiYiLin, Yi-JenYi-JenLinHo, Ping-RuiPing-RuiHoSHIH-CHIN YANG2025-08-142025-08-142025-06-15978166545776701972618https://www.scopus.com/record/display.uri?eid=2-s2.0-105011089495&origin=resultslisthttps://scholars.lib.ntu.edu.tw/handle/123456789/731398Bearings are critical components in rotating machinery. Their failures can lead to equipment downtime or even severe accidents. Traditional bearing fault diagnosis methods rely on vibration-based sensors. These sensors suffer from high costs and susceptibility to environmental noise. Under this effect, this paper proposes a bearing fault diagnostic method based on Motor Current Signature Analysis (MCSA). On the basis, inherent current signals of permanent magnet synchronous motors (PMSMs) are utilized for fault detection. The proposed current-based bearing diagnosis develops the three-phase current synchronization and Fast Fourier Transform (FFT) for online calculation of fault characteristic frequencies. The diagnosis accuracy is validated on a small-scale PMSM test platform. Experimental results demonstrate that the proposed method can accurately diagnose bearing faults under low load and low speed conditions. More importantly, all the diagnostic algorithms can be implemented by a 32-bit microcontroller at low cost.falseearly fault diagnosis (EBD)motor current signature analysis (MCSA)prognostic management system (PMS)rolling bearing cracks[SDGs]SDG7[SDGs]SDG13conference paper10.1109/IAS62731.2025.110616192-s2.0-105011089495