摘要
电力数据中存在着大量噪声,导致正常数据与扰动数据难以分离,因此提出了基于多级中值滤波的电力扰动数据分离建模方法。选定多级中值滤波窗口,利用该窗口对电力数据进行滤波处理,保留电力数据的细节特征,实现噪声抑制。分析扰动源数据激发的问题,结合干扰源的延迟时间编码搭建电力扰动数据感知模型。根据电力扰动数据感知结果以及稀疏约束范数项,构建电力扰动数据分离模型,实现正常数据和扰动数据的分离。由实验结果可知,该方法的电力扰动数据分离效果好,且最大信噪比为41 dB,达到了研究预期。
There are a lot of noises in power data,which makes it difficult to separate normal data from disturbance data.Therefore,a modeling method of power disturbance data separation based on multilevel median filter is proposed.The multi-level median filtering window is selected to filter the power data,retain the details of the power data,and achieve noise suppression.The positive problem of disturbance source data excitation is analyzed,and the power disturbance data perception model is built based on the delay time coding of disturbance source.According to the sensing results of power disturbance data and sparse constraint norm terms,a power disturbance data separation model is built to separate normal data from disturbance data.From the experimental results,it can be seen that the power disturbance data separation effect of this method is good,and the maximum signal-to-noise ratio is 41 dB,which meets the research expectation.
作者
张征凯
黄道友
徐晓波
ZHANG Zhengkai;HUANG Daoyou;XU Xiaobo(State Grid Anhui Electric Power Co.,Ltd.,Hefei 230022,China;Anhui Mingsheng HENGZHUO Technology Co.,Ltd.,Hefei 230094,China)
出处
《电子设计工程》
2024年第14期73-76,81,共5页
Electronic Design Engineering
关键词
多级中值滤波
电力扰动
数据分离
稀疏约束
multilevel median filter
power disturbance
data separation
sparse constraint
