摘要
铝绞线断裂和表面损伤将逐渐发展为对导线和整个输电线路的不可逆破坏,甚至导致严重停电。为了发现潜在的断股和损伤故障,防止其进一步恶化,提出了一种输电线路无人机检测导线断裂和表面缺陷的识别方法,首先通过无人机图像采集系统获取导线图像,然后通过灰度方差归一化(gray variance normalization, GVN)增强处理后的自适应阈值分割提取导线区域。导线破裂是由灰度分布曲线的方波变换(square wave transformation, SWT)检测的。同时,利用导线区域GVN图像的投影算法识别导线表面缺陷。最后通过计算断股数,过滤可疑缺陷,得到故障诊断结果。通过一系列实验对该技术的性能进行了分析,结果表明了所提方法的有效性。
The breakage and surface damage of aluminum strand will gradually develop into irreversible damage to the conductor and the whole transmission line,and even lead to serious power failure.In order to find the potential broken strand and damage faults and prevent them from further deterioration,a recognition method for detecting conductor fracture and surface defects by unmanned aerial vehicle(UAV)is proposed.Firstly,the conductor image is obtained by UAV image acquisition system,and then the gray variance is normalized.The conductor region is extracted by adaptive threshold segmentation after gray variance normalization(GVN)enhancement.Then,the conductor rupture is detected by square wave transformation(SWT)of gray distribution curve.At the same time,the GVN image projection algorithm is used to identify the surface defects.Finally,the fault diagnosis results can be obtained by calculating the number of broken strands and filtering the suspicious defects.Through a series of experiments,the performance of the technology is analyzed,and the results show the effectiveness of the proposed method.
作者
冯河清
鹿可可
文裕钧
范岩
FENG Heqing;LU Keke;WEN Yujun;FAN Yan(Guangxi Power Grid Co.,Ltd.,Nanning Guangxi 530000,China)
出处
《传感技术学报》
CAS
CSCD
北大核心
2022年第8期1123-1130,共8页
Chinese Journal of Sensors and Actuators
基金
基于三维激光点云的无人机智能巡检研究及示范应用(GXKJXM20190200)。
关键词
输电线路
缺陷识别
图像处理
传输线
无人机检验
transmission line
defect identification
image processing
UAV inspection
