摘要
随着我国减排降碳战略的实施,绿色能源大规模发展。光伏发电作为最优质的绿色能源之一,光伏发电技术发展尤为迅速。光伏组件的故障识别作为光伏发电技术中关键一环,特别提出一种基于无人机拍摄的红外图片结合深度学习算法Faster RCNN进行光伏板上的热斑识别。在原始Faster RCNN的基础上,结合改进特征提取网络、双线性插值面积积分固定建议框尺寸,得到热斑故障识别模型。
With the implementation of China’s carbon reduction strategy,green energy has developed on a large scale.As one of the best green energy,photovoltaic power generation technology has developed particularly rapidly.As a key part of photovoltaic power generation technology,fault identification of photovoltaic modules is proposed in this paper,which is based on infrared images taken by UAV and fast RCNN.Based on the original fast RCNN,combined with improved feature extraction network and bilinear interpolation area integral to fix the recommended frame size,the hot spot fault identification model is obtained.
作者
王洛南
白雪峰
郄慧广
刘东
刘泽
Wang Luo-nan;Bai Xue-feng;Qie Hui-guang;Liu Dong;Liu Ze
出处
《电力系统装备》
2021年第24期195-196,共2页
Electric Power System Equipment
