I. Introduction
Problems such as line insulation aging and grounding faults often lead to the occurrence of electric leakage, which poses a great threat to personal health and property safety. Therefore, in low-voltage power grids, in order to effectively prevent leakage fires and personal electric shock accidents, leakage current protection devices are widely used [1], [2]. At present, most of the leakage protection devices use the root mean square (RMS) value of the leakage current to judge the leakage [3]. This kind of leakage protection system lacks the ability to identify the type of electric shock and cannot judge whether it is a biological body shock. Therefore, in order to ensure personal safety, it is necessary to develop an adaptive leakage protection device that can accurately identify biological electric shocks [4].