摘要
传统的变压器呼吸器缺少检测手段，无法实现远程监测，必须通过运行工作人员定期巡视确定其状态．本文介绍了一种基于图像处理的变压器呼吸器自动检测方法，借助于智能巡检机器人现场拍摄并传回的变压器呼吸器装置图像，利用数字图像处理技术对该图片进行颜色直方图统计分析，并进行决策判断是否需要更换变色硅胶．这种检测方法不仅对变压器呼吸器的构造没有任何要求，同时也不需要任何传感器装置，因此大大降低了变压器呼吸器的检测成本，既方便又实用．
The traditional transformer respirator lacks the detection means, cannot realize the long-distance monitoring, and must carry on the regular inspection through the running staff to determine its state. This paper introduces a method of automatic detection of trans- former respirator based on image processing with the aid of the intelligent inspection robot on-site shooting, returning the transformer respirator device image, using the digital image processing technology for color histogram analysis of the picture, and last making decisions to determine whether the need to replace the discoloration of silica gel. This method not only does not require any requirement for the structure of the transformer respirator, but also does not require any sensor device, thus greatly reducing the detection cost of the transformer respirator, both convenient and practical.
作者
辛明勇
肖小兵
刘斌
林呈辉
熊金梅
袁熹
苏功闯
XIN Ming-yong;XIAO Xiao-bing;LIU Bin;LIN Cheng-hui;XIONG Jin-mei;YUAN Xi;SU Gong-chuang(Guizhou Electric Power Research Institute of Guizhou Power Grid Co Ltd, Guiyang 550002, China;Wuhan Zhongyuan Huadian Science ＆ Technology Co Ltd, Wuhan 430074, China)
出处
《数学的实践与认识》
北大核心
2018年第2期127-133,共7页
Mathematics in Practice and Theory
关键词
变压器呼吸器
图像处理
颜色直方图
远程监控
transformer respirator
image processing
color histogram
remote monitoring
