基于压缩感知估计行波自然频率的输电线路故障定位方法研究
发布时间：2019-03-04 11:53
【摘要】：提出了一种基于压缩感知估计行波自然频率的输电线路故障定位方法,能够准确估计多次行波自然频率,具有较高的故障定位精度。针对故障行波信号的频域特征设计了新的过完备字典,使行波信号能够被有效地稀疏表示。进行故障定位时,首先用小波模极大值方法确定故障行波信号的频率边界,并采用FIR滤波进行预处理滤除低频干扰成分,然后将其变换到频域使其能够在过完备字典上稀疏表示。在此基础上,该文利用改进的基于Dice系数的OMP算法(DOMP算法)对故障行波的频域信号进行重构,精确辨识行波信号的多次自然频率值,最终结合反射角和波速实现准确地故障定位。通过仿真分析验证了字典设计结合改进的DOMP算法的方法具有较高的定位精度和可靠性。
[Abstract]:This paper presents a fault location method for transmission line based on compression sensing to estimate the natural frequency of traveling wave, which can accurately estimate the natural frequency of multiple traveling waves and has high fault location accuracy. A new over-complete dictionary for the frequency domain characteristics of fault traveling wave signals is designed to make the traveling wave signals can be represented sparsely effectively. When the fault is located, the wavelet modulus maximum method is used to determine the frequency boundary of the fault traveling wave signal, and the FIR filter is used to pre-filter the low-frequency interference components. Then it is transformed into frequency domain so that it can be sparsely represented in an over-complete dictionary. On this basis, the improved OMP algorithm based on Dice coefficient (DOMP) is used to reconstruct the frequency domain signal of the fault traveling wave, and the multiple natural frequency values of the traveling wave signal are accurately identified. Finally, the fault location is realized by combining reflection angle and wave velocity. The simulation results show that the dictionary design combined with the improved DOMP algorithm has high positioning accuracy and reliability.
【作者单位】： 东北电力大学信息工程学院;东北电力大学电气工程学院;
【基金】：国家自然科学基金(51307020) 吉林省科技厅项目(20160520089JH)资助
【分类号】：TM75
，
本文编号：2434247
[Abstract]:This paper presents a fault location method for transmission line based on compression sensing to estimate the natural frequency of traveling wave, which can accurately estimate the natural frequency of multiple traveling waves and has high fault location accuracy. A new over-complete dictionary for the frequency domain characteristics of fault traveling wave signals is designed to make the traveling wave signals can be represented sparsely effectively. When the fault is located, the wavelet modulus maximum method is used to determine the frequency boundary of the fault traveling wave signal, and the FIR filter is used to pre-filter the low-frequency interference components. Then it is transformed into frequency domain so that it can be sparsely represented in an over-complete dictionary. On this basis, the improved OMP algorithm based on Dice coefficient (DOMP) is used to reconstruct the frequency domain signal of the fault traveling wave, and the multiple natural frequency values of the traveling wave signal are accurately identified. Finally, the fault location is realized by combining reflection angle and wave velocity. The simulation results show that the dictionary design combined with the improved DOMP algorithm has high positioning accuracy and reliability.
【作者单位】： 东北电力大学信息工程学院;东北电力大学电气工程学院;
【基金】：国家自然科学基金(51307020) 吉林省科技厅项目(20160520089JH)资助
【分类号】：TM75
，
本文编号：2434247
本文链接：https://www.wllwen.com/kejilunwen/dianlidianqilunwen/2434247.html
