机车电传动 (May 2024)
Identification of partial discharges in cable terminals of high-speed EMUs based on fuzzy C-means clustering
Abstract
As an effective means to diagnose the insulation status of on-board cable terminals, partial discharge detection faces strong interference in the actual operating environment of trains. To address this issue, this paper proposed a strategy for separating partial discharge pulses of on-board cable terminals based on waveform parameter analysis and fuzzy C-means clustering. A partial discharge test platform was built in the laboratory, and high-frequency current transducers (HFCT) were used to acquire partial discharge signals and the typical pulse interference signals from cable terminals. By performing envelope analysis on individual pulses, three parameters of the pulses were extracted as the feature vectors. Subsequently, fuzzy C-means clustering was employed to separate the partial discharge signals from the pulse interference signals. The experimental results demonstrate that the proposed method can effectively separate partial discharge signals from pulse interference signals, reducing the impact of pulse interference on partial discharge detection, and is of some significance in improving the accuracy of assessing the insulation status of the on-board cable terminals through partial discharge means.
