摘要
为解决传统电力系统巡检效率慢、巡检精度差的问题,提出基于云端控制协同的无人机自动驾驶智能巡检技术。搭建云端控制协同平台,云服务器通过内网接入智能巡检系统,设计无人机巡检多线程定位模块,通过PC端接收图像坐标模块的坐标信息,显示模块通过云平台实现数据共享;划分图像中心点坐标并标记,利用高斯核函数提升像素分布集中程度;最后拟定无人机悬停点,通过云端控制协同平台判断所获得的巡检路线图像是否存在异常,实现自动驾驶智能巡检作业。实验证明:所提方法能够精确识别初始图像像素点为574×574的图像的边缘;巡检航线水平误差最大值为0.5 m,垂直误差最大值为1 m,且巡检航线的运动未出现偏离或者抖动现象。为实现电网可靠、安全的运行提供了一定的技术支持。
In order to solve the problems of slow inspection efficiency and poor inspection accuracy of traditional power systems,an intelligent inspection technology for UAV autonomous driving based on cloud control coordination is proposed.A cloud control collaboration platform is built.The cloud server is connected to the intelligent inspection system through the intranet,and the UAV inspection multi-thread positioning module is designed,the coordinate information of the image coordinate module is received through the PC,and the display module realizes data sharing through the cloud platform.The coordinates of the center point of the image are divided and marked,and the Gaussian kernel function is used to improve the concentration of pixel distribution.Finally,the hovering point of the drone is drawn up,and the inspection route image obtained through the cloud control collaborative platform is detected to judge whether there is an abnormality,so as to realize the intelligent inspection operation of autonomous driving.Experiments have proved that the proposed method can accurately identify the edge of an image with an initial image pixel of 574*574,the maximum horizontal error is 0.5m,and the maximum vertical error is 1m.There is no deviation or jitter in the movement of the inspection route,which can provide certain technical support for the reliable and safe operation of the power grid.
作者
张虎
任欣元
张纪宾
刘岚
李彬
ZHANG Hu;REN Xinyuan;ZHANG Jibin;LIU Lan;LI Bin(Foshan Power Supply Bureau of Guangdong Power Grid Co.,Ltd.,Foshan 528000,Guangdong,China;China Southern Power Grid Digital Power Grid Research Institute Co.,Ltd.,Guangzhou 510000,Guangdong,China)
出处
《电网与清洁能源》
北大核心
2022年第5期42-48,共7页
Power System and Clean Energy
基金
南方电网公司科技项目(030600KK51200001)。
关键词
云端控制
无人机
自动驾驶
智能巡检
图像识别
cloud control
UAV
automatic driving
intelligent patrol inspection
image recognition
