Application of Short-time Fourier Transform in Feeder Fault Detection of Flexible Multi-state Switch

Abstract

Flexible multi-state switch (FMSS) is widely used in active distribution networks. However, the medium voltage distribution network that includes FMSS requires more accurate fault detection method. To solve this problem, the application of short-time Fourier transform (STFT) in the design of FMSS fault detection methods is studied. The data of power spectrum density in the STFT spectrogram is used to define the Fault Index, Λ. The A can be used to accurately detect the fault type of FMSS in a relatively short time. In order to verify the effectiveness of this fault detection method, this paper builds a control structure for each port based on the FMSS topology and mathematical model. The FMSS model is simulated in MATLAB/Simulink, and the fault signal is analysed by STFT in MATLAB. The results show that the proposed fault detection method can accurately detect the fault type. The STFT parameters have a great influence on the fault detection result, which is studied in the parameters influence analysis.