A Method of Quickly Detecting Short-time Disturbance in Grid Voltage

Abstract

The rapid detection of grid voltage faults is crucial to achieve grid voltage fault ride-through. However, conventional grid voltage detection methods, including Fourier transform, dq axis transform, wavelet transform, and voltage peak methods, generally yield varying degrees of distortions in the detection results of short-time disturbances in grid voltage. To address this, this paper proposes an approach for quickly detecting short-time disturbances in grid voltage. Leveraging the distinctive high harmonic content present in grids, the suggested strategy involves a transformation of grid voltage calculated using the Fourier transform method, known for its robust anti-interference performance, into high-frequency signals through coordinate conversion. The integration of the rolling sliding window method addresses the extended calculation time issue associated with the Fourier transform, enabling the detection of grid voltage disturbance signals within a single high-frequency signal cycle using this method. The simulation experiment results demonstrated the wide applicability of the proposed method compared with other methods. It is not only effective for short-time grid voltage disturbances, but also suitable for long-time grid voltage disturbances. Regardless of whether high-frequency harmonics are present or absent in the grid voltage, this method allows for the rapid detection of grid voltage disturbances (within about 2 ms). Moreover, its accuracy in detecting grid voltage amplitudes aligns with requirements in engineering applications.

Keywords

• grid detection
• short-time disturbance
• fourier transform
• rolling sliding window