基于改进小波分析的电弧故障高频信号识别  

Arc fault high frequency signal identification based on improved wavelet analysis

作　　者：陈磊 高迪雅 CHEN Lei;GAO Diya(Chongqing Fuling Electric Power Industrial Co.,Ltd.,Chongqing 408000,China)

机构地区：[1]重庆涪陵电力实业股份有限公司,重庆408000

出　　处：《电子设计工程》2024年第6期122-126,共5页Electronic Design Engineering

摘　　要：电力故障监测范围较广,电弧故障高频信号识别准确率不高,因此引入改进小波分析技术,设计电弧故障高频信号识别方法。按照故障类型将电弧故障分为串行电弧、并行电弧、接地电弧。检验电弧高频信号的正则性,确定平方函数的连续小波变换结果,通过离散化处理分析故障特征。根据特征提取结果利用小波模极大值理论完成信号降噪处理,采用分级保留识别故障波形,通过电路信号分析实现故障高频信号识别。实验结果表明,该方法对时刻突变点故障识别准确率高达99.99%,对连续间断点故障识别准确率高达99.2%,故障信号识别效果好。The power fault monitoring range is wide,and the accuracy of arc fault high frequency signal recognition is not high.Therefore,the improved wavelet analysis technology is introduced to design the arc fault high frequency signal recognition method.According to the fault type,the arc fault is divided into serial arc,parallel arc and grounding arc.Check the regularity of arc high frequency signal,determine the continuous wavelet transform result of square function,and analyze the fault characteristics through discrete processing.According to the feature extraction results,the wavelet modulus maximum theory is used to complete the signal noise reduction processing,the hierarchical retention method is used to identify the fault waveform,and the circuit signal analysis is used to identify the fault high frequency signal.The experimental results show that the method has a high accuracy of 99.99%for fault identification of time mutation point and 99.2%for fault identification of continuous breakpoint,and the recognition effect of fault signal is good.

关 键 词：改进小波分析 电弧故障 故障信号 高频信号 信号识别 

分 类 号：TN-9[电子电信]