带电校验用钳形电流互感器二次信号降噪研究  

Study on Noise Reduction of Secondary Signal of Clamp Current Transformer for Live Calibration

作　　者：王钰力 吴田 王申华 陈将宏 WANG Yu-li;WU Tian;WANG Shen-hua;CHEN Jiang-hong(College of Electrical Engineering&New Energy,China Three Gorges University,Yichang Hubei 443000,China;State Grid Zhejiang Wuyi County Power Supply Co.,Ltd.,Jinhua Zhejiang 321200,China)

机构地区：[1]三峡大学电气与新能源学院,湖北宜昌443000 [2]国网浙江省电力有限公司武义县供电公司,浙江金华321200

出　　处：《计算机仿真》2024年第12期395-401,共7页Computer Simulation

基　　金：国家自然科学基金(51807110)。

摘　　要：随着大量分布式能源并入配电网,定期对互感器的误差特性进行校验极为重要,现场复杂电磁环境和噪声严重影响带电校验用钳形电流互感器的角差和比差的测量精度。为了解决强白噪声和随机脉冲干扰对带电校验用钳形互感器二次弱信号的影响,首先将输出信号进行集合经验模态分解(Ensemble Empirical Mode Decomposition,EEMD),选择适当的固有模态函数IMF(Intrinsic Mode Function),构造独立成分分析(independent component analysis,ICA)的虚拟噪声通道;采用MASM累计均值(mean of standardized accumulated modes,MSAM)提取IMF中的高频和低频分量,高频分量通过小波变换(Wavelet transform,WT)、低频分量通过ICA进行降噪处理。通过仿真信号与试验数据降噪比较,并采用信噪比、均方误差、波形相似系数等指标对降噪效果进行评价,结果表明,所提方法与ICA-Wavelet、DFT相比,信噪比和波形相似系数分别提高了28.5%、1.1%,且均方误差降低了30.8%,有更好的降噪效果。As a large number of distributed energy resources are integrated into the distribution network,it is very important to check the error characteristics of CT regularly.The complex electromagnetic environment and noise on site seriously affect the measurement accuracy of Angle difference and specific difference of CT for on-line calibration.In order to solve the influence of strong white noise and random pulse interference on the secondary weak signal of the on-line calibration tare transformer,the output signal is first decomposed by Ensemble Empirical Mode Decomposition(EEMD).The Intrinsic Mode Function(IMF)was selected to construct the virtual noise channel of independent component analysis(ICA).High frequency and low frequency accumulated modes in IMF are extracted by means of MASM accumulated mean value accumulated Modes(MSAM).High frequency accumulated modes are denoised by Wavelet transform(WT)and low frequency accumulated modes by ICA.By comparing the simulation signal with the test data,and using the signal-to-noise ratio,mean square error,waveform similarity coefficient and other indicators to evaluate the noise reduction effect.The results show that,compared with ICA-Wavelet and DFT,the proposed method has much better noise reduction effect.The signal-to-noise ratio and waveform similarity coefficient increase 28.5%and 1.1%,respectively,and the mean square error decreases 30.8%.

关 键 词：钳形电流互感器 集合经验模态分解 小波变换 独立成分分析 

分 类 号：TM743[电气工程—电力系统及自动化] TP301.6[自动化与计算机技术—计算机系统结构]