基于椭圆域分割的触电电流混沌检测方法研究

摘要

It is difficult to exactly detect the electric shock current signal of the human body from the leakage current in the residual current protection of low voltage power network.The chaotic system is sensitive to initial conditions and immune noise.Chaotic detection method based on elliptic domain segmentation method is proposed.Elliptic domain segmentation line is introduced in the two-dimensional phase track diagram of chaos system.The state of the phase trajectory passing through the elliptic domain segmentation line is indicated by the high and low level of the output of the elliptic domain segmentation.According to the number of the low level of the output,the state of the chaotic system is determined quantitatively,so as to realize the extraction of the electric shock signal.According to the relationship between the detecting signal phase and chaos system state,the initial phase of the detecting signal is confirmed and corrected to improve the accuracy of the method.The detection results of 195 groups of electric shock signal show that the detection error of the phase corrected before and after is reduced from 32.5％ to 4.1％,SNR range detected is increased from [0,-50 dB] to [0,-63 dB].Simulation results show that this method can detect weak electric shock current in the summation leakage current containing strong noise.The detecting method is of a certain reference value for developing new generation residual current operation devices.%针对从低压电网的总剩余电流中提取触电支路电流的难题,利用混沌系统对初始条件敏感及对噪声免疫的特性,提出基于椭圆域分割法的触电电流混沌检测方法.在混沌系统的二维相轨迹图中引入椭圆域分割线,通过其输出的高低电平表示相轨迹穿越椭圆域分割线的状态,根据输出的低电平次数定量判别混沌系统所处的状态,从而实现触电信号的提取.利用待测信号相位和系统状态存在的对应关系,确定待测信号初相位并修正,以提高该方法的检测精度.针对195组触电信号检测结果表明,相位修正前、后的检测误差从32.5％减小到4.1％,检测信噪比由[0,-50 dB]增加到[0,-63 dB],所提议的方法能够从包含强噪声的总剩余电流中检测出触电电流,为新型剩余电流保护装置的开发提供一定的参考价值.