부산대학교 도서관

The existing residual current protectors mostly use the total residual current or the effective value of the fault current as the action criterion, which is prone to malfunction or failure to operate. In order to improve the accuracy of residual current identification and avoid serious consequences caused by personal electric shock accidents, this paper proposes a low-voltage residual current signal feature extraction method based on wavelet packet energy ratio. First, build a biological electric shock experimental platform and collect residual current data. Secondly, the sliding time window is used to perform wavelet packet transformation on the signal to obtain the changes in wavelet packet energy and wavelet packet energy ratio in each frequency band of the signal. Finally, a calculation example is simulated and the characteristic values of electric shock faults are extracted. Experimental results show that the wavelet packet energy ratio of residual current can highlight the main characteristics of biological electric shock and can be used as an effective means to judge biological electric shock faults.