期刊论文

Research on high-voltage cable sheath faults identification based on Lissajous figure and relation analysis

Xia, Xiangyang(xia_xy@126.com)

中文

中南大学学报(自然科学版)

1672-7207

2020

51

4

989-997

10.11817/j.issn.1672-7207.2020.04.014

国家自然科学基金资助项目（51977014）~~；

本校为第一机构

针对高压输电电缆护层故障监测与识别的不足,提出一种基于利萨如图形及关联度分析的高压电缆护层故障识别方法。该方法包含基于利萨如图形的信号分析方法和基于关联度分析的模式识别方法,通过采集高压电缆首末端环流并构建二维利萨如图形,分析获取图形特征参数包括长轴长度、短轴长度、离心率、倾斜角及其变化率并结合采集的环流值作为输入特征向量,通过求解未知故障类型的输入特征向量与已知故障类型参数组成的样本空间向量的熵、熵权、关联系数等得到相应的关联度,实现故障的准确识别。最后以实际线路为参考搭建仿真模型并结合实例分析验证该方法的准确性。研究结果表明:该方法能较准确识别故...

Aiming at the inadequate of HV cable faults monitoring and identification, a method based on Lissajous figure and relation analysis was proposed, which included a signal analysis method based on Lissajous figure and a pattern recognition method based on grey correlation degree. By measuring two circulating currents in a coaxial cable simultaneously, a two-dimensional Lissajous figure was constructed to obtain graphical feature parameters including changing rate of long axis, short axis length, eccentricity length and tilt angle that were used a...