Real-Time Detection Method for Saturation Speed of Air-Core Coil Current Transformer

Abstract

Traditional methods generally rely on a fixed threshold to determine whether the current transformer is saturated. However, this threshold based judgment method is difficult to adapt to different working conditions and changes in current waveforms. In view of the different characteristics of current transformer and the diversity of working environment, it may not be accurate to judge its saturation state only by relying on a fixed threshold. In order to improve the efficiency of current transformer speed saturation detection and reduce the error, a new type of air core coil current transformer speed saturation real-time detection method is proposed. Based on the working and current characteristics of the fast saturation current transformer, the discrete Fourier algorithm is used to suppress power frequency interference signals, with the fundamental frequencies set to 45 Hz and 55 Hz respectively, in order to separate the signal from power frequency interference and reduce the interference intensity of the power frequency interference signal; Identify all faults and saturation points, and extract effective data segments. The wavelet transform characteristics of the secondary current waveform of current transformer in saturated and unsaturated states are deeply studied. By analyzing the changes of different operating conditions and current waveforms, more discriminant features are extracted by using the characteristics of wavelet transform, so as to realize the real-time and accurate detection of saturation state. The results indicate that after adopting the proposed method, the highest accuracy of data segment extraction reached 97%; The maximum detection error is only 4%. The detection time is only 1.31 s. This indicates that the proposed method improves the detection performance and is practical.