针对无人机航拍光伏组件红外热图像中热斑信号微弱、特征模糊以及自动化巡检存在较高漏检率和误检率的问题,该文基于深度学习模型YOLOv5进行优化改进,以提升其对微弱红外热斑的检测能力。首先,通过添加混合特征模块（HFM）,进一步增强网络的特征提取能力,有效提高检测精度。其次,在YOLOv5的主干网络后使用层级聚合注入网络（HAI）,使卷积核能得到全局特征信息,从而提高检测精度。最后,引入WIoU损失函数优化原有的边界框损失,提高了困难样本的检测速度。实验结果显示,改进后的模型mAP达到80.3%,相比原模型提升了5个百分点。
Abstract
To address the challenges of weak hotspot signals and blurred features in infrared images of photovoltaic modules captured by drones, as well as the high miss and false detection rates in automated inspections, this paper presents an optimized version of the YOLOv5 deep learning model to enhance its ability to detect faint infrared hotspots. First, by incorporating a hybrid feature module （HFM）, the network's feature extraction capability is further enhanced, leading to a significant improvement in detection accuracy. Additionally, a hierarchical aggregation Injection （HAI） network is employed after the backbone of YOLOv5, enabling the convolutional kernels to capture global feature information, thereby enforcing high detection precision. Finally, the WIoU loss function is introduced to improve the original bounding box loss and increase the detection speed of difficult samples. The experimental results show that the mAP of the improved model reaches 80.3%, which is an improvement of 5 percentage points compared to the original model.
关键词
光伏组件 /
深度学习 /
目标检测 /
图像处理 /
无人机巡检 /
红外热成像
Key words
photovoltaic modules /
deep learning /
object detection /
image processing /
UAV inspection /
infrared thermal imaging
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 刘喜梅, 崔阳, 董建明, 等. 基于风电机组可靠性的风电场平准化成本模型研究[J]. 太阳能学报, 2022, 43(10): 304-311.
LIU X M, CUI Y, DONG J M, et al.Research on levelized cost model of wind farms based on wind turbine reliability[J]. Acta energiae solaris sinica, 2022, 43(10): 304-311.
[2] 周亚男. 光伏电站运维现状分析[J]. 太阳能, 2024(1): 12-19.
ZHOU Y N.Analysis on current situation of operation and maintenance of pv power stations[J]. Solar energy, 2024(1): 12-19.
[3] 卢和平, 陈昊, 白静芬, 等. 新型电力系统动态电能测量技术及其标准化研究[J]. 电网技术, 2024, 48(4): 1383-1395.
LU H P, CHEN H, BAI J F, et al.Research on dynamic energy measurement technology and standardization for new power system[J]. Power system technology, 2024, 48(4): 1383-1395.
[4] 孙海蓉, 周映杰, 张镇韬, 等. 基于改进自私羊群算法的光伏红外热图像热斑识别方法[J]. 中国电机工程学报, 2022, 42(24): 8942-8950.
SUN H R, ZHOU Y J, ZHANG Z T, et al.Hot spot recognition method of photovoltaic infrared thermal image based on improved selfish herd algorithm[J]. Proceedings of the CSEE, 2022, 42(24): 8942-8950.
[5] WANG Y D, ITAKO K, KUDOH T, et al.Voltage-based hot-spot detection method for PV string using projector[C]//2016 IEEE International Conference on Power and Renewable Energy (ICPRE). Shanghai, China, 2016: 570-574.
[6] 彭自然, 许怀顺, 肖伸平. 一种基于CatBoost优化的光伏阵列故障诊断模型[J]. 电子学报, 2024, 52(7): 2418-2428.
PENG Z R, XU H S, XIAO S P.A CatBoost optimization-based fault diagnosis model for photovoltaic arrays[J]. Acta electronica sinica, 2024, 52(7): 2418-2428.
[7] LI Y L, DING K, ZHANG J W, et al.A fault diagnosis method for photovoltaic arrays based on fault parameters identification[J]. Renewable energy, 2019, 143: 52-63.
[8] 刘恒, 马铭遥, 张志祥, 等. 热斑晶硅光伏组件I-V曲线分析及特征模拟研究[J]. 太阳能学报, 2021, 42(4): 239-246.
LIU H, MA M Y, ZHANG Z X, et al.Study on I-V curve analysis and characteristic simulation of silicon photovoltaic module with hot spot[J]. Acta energiae solaris sinica, 2021, 42(4): 239-246.
[9] 吕艳玲, 钟晨, 刘志鹏. 基于光伏组件参数辨识的故障诊断分析[J]. 上海交通大学学报, 2025, 59(9): 1383-1396.
LYU Y L, ZHONG C, LIU Z P.Fault diagnosis analysis based on photovoltaic module parameter identification[J]. Journal of Shanghai Jiaotong University, 2025, 59(9): 1383-1396.
[10] SUN T Y, XING H S, CAO S X, et al.A novel detection method for hot spots of photovoltaic (PV) panels using improved anchors and prediction heads of YOLOv5 network[J]. Energy reports, 2022, 8: 1219-1229.
[11] 赵波, 廖坤, 邓春宇, 等. 基于卷积神经学习的光伏板积灰状态识别与分析[J]. 中国电机工程学报, 2019, 39(23): 6981-6989.
ZHAO B, LIAO K, DENG C Y, et al.Image convolutional neural learning based image recognition and analysis method for dust on photovoltaic panel[J]. Proceedings of the CSEE, 2019, 39(23): 6981-6989.
[12] 王道累, 李超, 李明山, 等. 基于深度卷积神经网络的光伏组件热斑检测[J]. 太阳能学报, 2022, 43(1): 412-417.
WANG D L, LI C, LI M S, et al.Solar photovoltaic modules hot spot detection based on deep convolutional neural networks[J]. Acta energiae solaris sinica, 2022, 43(1): 412-417.
[13] 彭自然, 张颖清, 肖伸平. 基于YOLOv5的太阳电池表面缺陷检测[J]. 太阳能学报, 2024, 45(6): 368-375.
PENG Z R, ZHANG Y Q, XIAO S P, et al.Research on surface defect detection of solar cell with improved YOLOv5 algorithm[J]. Acta energiae solaris sinica, 2024, 45(6): 368-375.
[14] 王银, 高瑞泽, 李茂环, 等. 基于YOLOv5 LiteX算法的光伏组件缺陷检测[J]. 太阳能学报, 2023, 44(9): 101-108.
WANG Y, GAO R Z, LI M H, et al.Photovoltaic modules defect detection based on YOLOv5 LiteX algorithm[J]. Acta energiae solaris sinica, 2023, 44(9): 101-108.
[15] 王昊, 丁国斌, 杨家慧. 基于无人机机载AI模块的电力巡检技术[J]. 沈阳工业大学学报, 2024, 46(1): 49-53.
WANG H, DING G B, YANG J H, et al.et al. Power inspection technology based on UAV airborne AI module[J]. Journal of Shenyang University of Technology, 2024, 46(1): 49-53.
[16] 王道累, 姚从荣, 李超, 等. 光伏组件热斑效应智能化检测的综述及展望[J]. 太阳能学报, 2024, 45(5): 527-536.
WANG D L, YAO C R, LI C, et al.Overview and prospect of intelligent detection of hot spot effect of photovoltaic modules[J]. Acta energiae solaris sinica, 2024, 45(5): 527-536.
[17] 刘熹, 陈晨, 双丰. 基于改进YOLOv7-tiny的多种类绝缘子检测算法[J]. 仪器仪表学报, 2024, 45(9): 101-110.
LIU X, CHEN C, SHUANG F.Multi-type insulator detection algorithm based on improved YOLOv7-tiny[J]. Chinese journal of scientific instrument, 2024, 45(9): 101-110.
基金
国家重点研发计划（2019YFE0122600）; 湖南省教育厅重点科研项目（22A0423）; 湖南省自科科学基金（2023JJ60267; 2022JJ50073）
PDF(2132 KB)
