基于自适应最稀疏时频分析的电力线线夹段缺陷检测研究

Research on Overhead Wires in Clips Defect Detection Based on Adaptive Sparse Time-Frequency Analysis

电力钢绞线是电能输送的载体,电塔上的悬垂线夹是提拉电力钢绞线的关键部件。悬垂线夹结构具有半封闭特点,导致线夹内的钢绞线在受到风载、雨水侵蚀或雷击等外部作用后,更易产生较为隐蔽的缺陷。为有效检测线夹段架空线缆的状态,根据线缆的常见缺陷特征及导波特性,提出了一种基于ASTFA和信号时频熵的损伤识别方法。首先,对GJ100型钢绞线中的导波传播特性进行分析;其次,引入自适应最稀疏时频分析算法和时频熵构建损伤评估指标,并对有限元仿真信号进行指标计算,初步验证该指标的有效性;最后,实验室及现场的实验证明了该方法具有可行性。结果表明:(1)低频低阶纵向导波更适用于钢绞线检测,本文选取64 kHz的L(0,1)模态;(2)随着损伤程度的增加,损伤评估指标的值上升,且导波频率对指标值的影响有限;(3)根据实验划分了损伤程度评价范围,现场试验证明该指标较为可靠。

The transmission of electric energy relies on the power steel strand,and the suspension clamp on the electrical tower is the key component of the power steel strand.The structure of the suspension clamp has a semi-closed feature,which makes the steel strands in the clamp more likely to have more hidden defects after being exposed to external actions such as wind load,rain erosion or lightning strikes.In order to effectively detect the status of overhead cables in the clamp section,a damage identification method based on ASTFA and the time-frequency entropy of the signal is proposed according to the common defect characteristics and guided wave characteristics of the cable.First,analyze the propagation characteristics of the guided wave in the steel strand;secondly,introduce the self-adaptive sparsest time-frequency analysis algorithm and time-frequency entropy to construct the damage assessment index,and calculate the finite element simulation signal to initially verify the index Effectiveness;Finally,laboratory and field experiments have proved the feasibility of this method.The results show that:(1)low-frequency longitudinal guided waves are more suitable for steel strand detection and the L(0,1)mode of 64 kHz is selected in this paper;(2)as the degree of damage increases,the value of the damage assessment index increases,and the frequency of guided waves has a limited influence on the index value;(3)according to the experiment,the damage degree evaluation range is divided,and the field test proves that the index is more reliable.