Empirical Mode Decomposition Based DC Fault Detection Method in Multi-terminal DC System

摘要

The rapid rising rate of dc line current after the dc fault inception in the high voltage dc (HVDC) system proposes a high requirement on the response speed of dc fault detection methods. In this paper, a new detection method based on empirical mode decomposition (EMD) is proposed. Unlike traditional parameters including time and frequency, EMD tries to describe the signal by characteristic time scale (CTS) which is suited to analyze non-linear and non-stationary signals such as the fault current. The principle and procedure of this method is illustrated in detail. Generally, this method could detect the dc fault within 0.2 ms with enough discrimination to distinguish the fault line and fault-free lines on the 4-bus ring multi-terminal dc (MTDC) system in MATLAB/Simulink. In addition, this method could retain its validity free from the interference of noise and filtering. Moreover, a comparison of this EMD based method between wavelet transform (WT) and short time Fourier transform (STFT) is also made to show its fast detection speed.