摘要
提出了一种基于Faster R-CNN的图像处理方法,用于光伏并网变电站的运行故障检测。该方法首先对变电站的实时图像进行预处理,然后利用Faster R-CNN模型进行特征提取和区域建议,在标注的故障数据集上训练模型,实现了对光伏组件、连接线和设备故障的高效检测。实验结果表明,该方法能够准确地检测出多种类型的故障,显著提高了变电站运行的安全性和可靠性。
This study proposes an image processing method based on Faster R-CNN for fault detection in photovoltaic grid connected substations.This method first preprocesses the real-time images of the substation,and then uses the Faster RCNN model for feature extraction and region recommendation.The model is trained on the annotated fault dataset to achieve efficient detection of faults in photovoltaic modules,connecting lines,and equipment.The experimental results show that this method can accurately detect various types of faults,significantly improving the safety and reliability of substation operation.
作者
赵红伟
李朋
程振飞
ZHAO Hongwei;LI Peng;CHENG Zhenfei(Tongchuan Xiaguang New Energy Power Generation Co.,Ltd.,Tongchuan 727000,China;China Three Gorges New Energy(Group)Co.,Ltd.Shaanxi Branch,Xi′an 710076,China)
出处
《电工技术》
2025年第1期27-29,32,共4页
Electric Engineering
