基于深度学习的输电线路巡检系统的设计与实现  

Design and Implementation of Transmission Line Inspection System Based on Deep Learning

作　　者：李晋 王玲桃[1] 刘帅[1] 张朋朋 Li Jin;Wang Lingtao;Liu Shuai;Zhang Pengpeng(School of Electric Power and Architecture,Shanxi University,Taiyuan Shanxi 030006,China)

机构地区：[1]山西大学电力与建筑学院,山西太原030006

出　　处：《现代工业经济和信息化》2024年第12期116-120,243,共6页Modern Industrial Economy and Informationization

摘　　要：绝缘子作为输电线路上最重要元件之一,用来保持带电体与杆塔之间良好的绝缘。对电力系统的发展发挥着重要的作用,但因为绝缘子常年暴露于自然环境中,发生故障的概率较高。因此对绝缘子进行实时监控检测,避免因绝缘子破损威胁输电的安全稳定性具有十分重要的意义。随着计算机视觉的发展,目标检测算法已经被广泛应用于输电线路巡检中,采用当下检测效果比较好的YOLOv5s模型为基础,对收集的绝缘子数据集进行人工标注后进行训练与测试,基于Py Charm和Py Qt5设计了一个输电线路巡检系统。通过Qt中UI界面,完成了系统界面的布局与设计,可以实现数据源的选择、实时图像显示、检测操作、结果统计的功能。通过对模型YOLOv5s的调用,完成输电线路巡检系统的设计及实现。Insulators,as one of the most important components on transmission lines,are used to maintain good insulation between charged bodies and towers.It plays an important role in the development of the power system,but because insulators are exposed to the natural environment all year round,the probability of failure is high.Therefore,it is of great imp ortance to carry out real-time monitoring and detection of insulators to avoid threatening the safety and stability of power transmission due to insulator breakage.With the development of computer vision,target detection algorithms have been widely used in transmission line inspection,using the YOLOv5s model,which has a better detection effect nowadays,as the basis,the collected insulator dataset is manually labeled and then trained and tested,and then a transmission line inspection system is designed based on Py Charm and Py Qt5.Through the UI interface in Qt,the layout and design of the system interface is completed,which can realize the functions of data source selection,real-time image display,inspection operation and result statistics.Through the call of model YOLOv5s,the design and realization of transmission line inspection system is completed.

关 键 词：绝缘子破损 YOLOv5s PyCharm PyQt5 系统设计 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]