A New-Designed Biological Electric Shock Identification Method in Low-Voltage Distribution Network

• Published in: IEEE transactions on power delivery Vol. 38; no. 3; pp. 1558 - 1568
• Main Authors: Yang, Gengjie, Quan, Shengxin, Gao, Wei
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.06.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: adaptive startup threshold; biological electric shock(BES); Circuits; Covariance matrices; Distribution networks; Electric potential; Electric shock; Fibrillation; Identification methods; Kalman filters; Low voltage; Power system reliability; Residual current device(RCD); Sampling; Voltage; Waveforms; weighted sum of deviation
• ISSN: 0885-8977; 1937-4208
• DOI: 10.1109/TPWRD.2022.3217880

In general, residual current devices (RCDs) have problems such as protection dead-zone and difficulty in threshold setting. A new method for identification of biological electric shock (BES) in low-voltage distribution network based on threshold method is proposed. Firstly, the total residual current of the circuit is denoised by Kalman filter, and then two threshold methods are investigated to determine the electric shock (ES) event and type respectively. Specifically, the first threshold consists of the maximum and average value of the current changes in the previous period, which is an adaptive value of dynamic change. If the current sampling value exceeds the threshold for 10 times in total within 5 ms, an ES is considered to have occurred. Then considering that the amplitude of the waveform of the first three periods after BES has the characteristics of gradual changes, the sampling values of the three periods are recorded. The second threshold is a fixed threshold which is obtained by weighting the phase point changes corresponding to the second-period and the third-period waveforms, and then the specific ES types are distinguished. The proposed method is implemented on hardware devices and analyzed in various common ES situations. The results show that for the three cycles of waveforms collected after the occurrence of grounding or ES, the accuracy of this method is 97.84% and the recognition time is 2.07 ms. In addition, based on the analysis of the actual BES data, a simple digital model is proposed to simulate the actual biological response, and it can be of great help in the subsequent study of such problems.