基于VMD和柔性形态学的输电线路故障测距方法.docVIP

基于VMD和柔性形态学的输电线路故障测距方法 陈惠，（） 摘要：针对variational modal decomposition ,VMD)对信号进行分解，在分解过程中自适应的去除一部分噪声；然后，通过平均柔性形态学滤波器进一步的去除残余干扰；最后使用柔性形态边缘检测有效放大信号奇异点，再设定输出阈值减少噪声背景对信号突变点的影响，可以很清晰的得到故障发生时刻。实验表明此方法有很好的噪声鲁棒性，在抑制了噪声干扰的同时放大了行波特征信号，能够有效检测输电线路故障行波波头，获得更高的检测精度。 关键词： 中图分类号：TM933 文献标识码：B 文章编号：1001-1390（201）00-0000-00 Transmission line fault location method based on VMD and flexible morphology C Hui, Mo Fujiang, Xu Mengsu , Wang Peipei (School of Electrical and Information Engineering, Jiangsu University, Zhenjiang 212013, Jiangsu, China) Abstract: Aiming at the problem of fault location and signal interference in transmission line, a fault location method of transmission line combining VMD (variational modal decomposition) with soft mathematical morphology was put forward in this paper. Firstly,VMD is used to decompose noisy signals, and a part of noise is automatically filtered out in the decomposition process. Then, the residual interference is further removed by the average flexible morphological filter. Finally, the flexible shape edge detection is used to amplify the singular point of the signal, and then, the output threshold is set to reduce the influence of the noise background on the abrupt change of the signal. The experimental results show that the proposed method has good noise robustness, and can suppressthe noise interference and amplify the traveling wave characteristic signal. It can detect the fault line of the transmission line effectively and obtain higher detection accuracy. Keywords: VMD, flexible morphology, signal de-noising, traveling wave detection 0　引 言 变分模态分解(VMD)是这几年新出来的一种自适应信号分解方法,分解模态稳定性好，能够更好地反应信号的奇异性特征。柔性形态学是基于积分几何和随机集论建立起来的一种非线性数学方法，使用柔性形态腐蚀、膨胀的不同结合，有效去除干扰的同时可以在时域快速得到突变点时刻。将VMD与柔性形态学结合起来，在有效去除噪声的基础上还进一步的放大了信号突变点，实现故障行波波头的有效定位。 1变分模态分解[8-10] 变分模态分解是一种最新的完全非递归的信号分解方法，尤其是对采样和噪声处理更强大，对于一系列人工和真实数据，我们可以达到预期的实际分解结果。VMD的目标是一个输入信号剖分成多个离散的子信号（模式）uk, 每个子信号K是主要环绕着一个确定跟随分解的中心脉动wk。该方法与EMD所使用的循环筛分剥离的信号处理方法有差别，采用迭代寻找变分模型最优解来确定每个