摘要
故障电弧是引发电气火灾的主要原因,其对居民的生命财产安全造成了严重威胁。目前,小波变换对于故障电弧有很好的辨识效果。为研究故障电弧辨识方法中小波变换的小波基选取问题,本文搭建基于小波变换和神经网络的故障电弧辨识模型,通过对小波基进行理论分析和仿真实验对比,给出选取最优小波基的参考依据。结果表明,db4小波基对故障电弧辨识效果显著,而sym2小波基在保证较高辨识成功率的同时节省嵌入式处理器的算力消耗,从而在实际工程中有更好的应用价值。
The fault arc is the main cause of electrical fires,which seriously threatens the safety of residents’ life and property.At present,wavelet transform has a good identification effect for the fault arc.In order to study the wavelet basis selection problem of the wavelet transform in the fault arc identification method,a fault arc identification model based on wavelet transform and neural network is built.Through the theoretical analysis of the wavelet basis and the comparison of simulation experiments,suggested principles for selecting the optimal wavelet basis are given.Our research results show that the db4 wavelet basis has a significant effect on the fault arc identification,and the sym2 wavelet basis not only ensures a higher recognition success rate but also saves the computing power consumption of the embedded processor at the same time,thus has better application value in practical projects.
作者
黄伟翔
林秀清
李珊
梁朔
欧世锋
HUANG Weixiang;LIN Xiuqing;LI Shan;LIANG Shuo;OU Shifeng(Electric Power Research Institute of Guangxi Power Grid Co.,Ltd.,Guangxi Nanning 530023,China)
出处
《广西电力》
2020年第3期44-48,52,共6页
Guangxi Electric Power
基金
广西电网有限责任公司科技项目(GXKJXM20180377)。
关键词
故障电弧
小波变换
小波基函数
神经网络
fault arc
wavelet transform
wavelet basis function
neural network
