一种高压线异物检测算法  被引量:3

High-voltage line foreign object detection algorithm

在线阅读下载全文

作  者:张凡[1,2] 范亚雷 刘文达 蔡涛 ZHANG Fan;FAN Yalei;LIU Wenda;CAI Tao(Hubei Collaborative Innovation Center for High Efficiency Utilization of Solar Energy,Hubei University of Technology,Wuhan 430068,China;Hubei Power Grid Intelligent Control and Equipment Engineering Technology Research Center,Hubei University of Technology,Wuhan 430068,China;Hubei Key Laboratory for High Efficiency Utilization of Solar Energy and Operation Control of Energy Storage,Hubei University of Technology,Wuhan 430068,China)

机构地区:[1]湖北工业大学太阳能高效利用湖北省协同创新中心,湖北武汉430068 [2]湖北工业大学湖北省电网智能控制与装备工程技术研究中心,湖北武汉430068 [3]湖北工业大学太阳能高效利用及储能运行控制湖北省重点实验室,湖北武汉430068

出  处:《现代电子技术》2020年第8期36-40,共5页Modern Electronics Technique

基  金:国家自然科学基金(11605051);湖北省高等学校优秀中青年科技创新团队计划项目(T201805).

摘  要:高压线路异物缠绕极易引发区域大面积停电,导致停运事故,智能异物检测算法是未来智能电力巡检系统的关键模块之一。通过整理近年人工处理高压线异物时备案的原始图片数据,对数据进行规范化处理和人工标注,建立"高压线异物检测数据集",研究一种高压线异物检测算法,该算法基于SSD模型,并有针对性地对数据集使用K-means聚类的方法获取先验框参数,以此替代默认值。实验结果在测试集上的平均准确率为86.69%,定位准确率为69.2%,表明所提算法可有效地对高压线上的异物进行检测和定位。The high-voltage line foreign object winding is easy to cause large-area regional blackout,resulting in the outage accidents. The intelligent foreign object detection algorithm is one of the key modules of the future intelligent power inspection system. The original picture data recorded in the manual processing of high-voltage line foreign objects in recent years is arranged,which is normalized and labeled manually,and the "high-voltage line foreign object detection data set" is established.A high-voltage line foreign object detection algorithm is studied. The algorithm is based on the SSD model,and the K-means clustering method is used to obtain the prior box parameters of the data set,so as to replace the default value. The average accuracy of the experimental results on the testing set is 86.69%,and the positioning accuracy is 69.2%. The algorithm can detect and locate foreign objects on the high-voltage line effectively.

关 键 词:异物检测 高压线 数据集构建 参数设置 模型训练 结果分析 

分 类 号:TN345-34[电子电信—物理电子学] TP391.4[自动化与计算机技术—计算机应用技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象