RECOGNITION AND CLASSIFICATION OF GRAYSCALE PHOTOVOLTAIC IMAGES USING CONVOLUTIONAL NEURAL NETWORKS FUSED WITH PERSISTENT HOMOLOGY

Sun Hairong; Tang Zhenchao; Zhang Hongwei; Zhou Lihui

doi:10.19912/j.0254-0096.tynxb.2024-0084

PDF(2303 KB)

Acta Energiae Solaris Sinica ›› 2025, Vol. 46 ›› Issue (6) : 321-328. DOI: 10.19912/j.0254-0096.tynxb.2024-0084

RECOGNITION AND CLASSIFICATION OF GRAYSCALE PHOTOVOLTAIC IMAGES USING CONVOLUTIONAL NEURAL NETWORKS FUSED WITH PERSISTENT HOMOLOGY

Sun Hairong^1,2, Tang Zhenchao^1,2, Zhang Hongwei^1,2, Zhou Lihui³

Author information +

History +

Abstract

Aiming at the low accuracy and high computational complexity of convolutional neural networks in recognizing and classifying photovoltaic infrared hot spot images, as well as the difficulty in identifying the hot spot features on photovoltaic infrared images, an algorithm based on persistent homology is proposed to extract topological features from grayscale photovoltaic hot spot images. Firstly, the photovoltaic infrared hot spot image is converted to grayscale. Then, persistent homology calculation is performed on the grayscale image to obtain a barcode, from which the topological features are extracted to compose a new image. Finally, a convolutional neural network is employed to recognize and classify the new image. The experimental results demonstrate that the grayscale photovoltaic infrared image is a single-channel image, resulting in lower computational complexity. The extracted topological features of the photovoltaic infrared hot spot image are easier to identify and classify, leading to higher accuracy.

Key words

feature extraction / convolutional neural network / persistent homology / topological data analysis / topological feature / identification and classification

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

Sun Hairong, Tang Zhenchao, Zhang Hongwei, Zhou Lihui. RECOGNITION AND CLASSIFICATION OF GRAYSCALE PHOTOVOLTAIC IMAGES USING CONVOLUTIONAL NEURAL NETWORKS FUSED WITH PERSISTENT HOMOLOGY[J]. Acta Energiae Solaris Sinica. 2025, 46(6): 321-328 https://doi.org/10.19912/j.0254-0096.tynxb.2024-0084

References

[1] 张智刚, 康重庆. 碳中和目标下构建新型电力系统的挑战与展望[J]. 中国电机工程学报, 2022, 42(8): 2806-2819.
ZHANG Z G, KANG C Q.Challenges and prospects for constructing the new-type power system towards a carbon neutrality future[J]. Proceedings of the CSEE, 2022, 42(8): 2806-2819.
[2] 王道累, 姚勇, 张世恒等. 基于红外热图像的光伏组件热斑深度学习检测方法[J]. 中国电机工程学报,2023, 43(24): 9608-9616.
WANG D L, YAO Y, ZHANG S H et al. Deep learning detection method of photovoltaic module hot spot based on infrared thermal image[J]. Proceedings of the CSEE,2023, 43(24): 9608-9616.
[3] 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): 8942-8951.
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): 8942-8951.
[4] 唐圣学, 邢明, 陈丽, 等. 太阳电池组串热斑防治策略研究[J]. 太阳能学报,2022,43(4):226-235.
TANG S X, XING M, CHEN L,et al.Study on suppressing strategy of hot spot in solar cellseries[J]. Acta energiae solaris sinica, 2022, 43(4): 226-235.
[5] 周颖, 毛立, 张燕, 等. 改进的CNN在太阳能电池板缺陷检测中的应用[J]. 计算机仿真, 2020, 37(3): 458-463, 474.
ZHOU Y, MAO L, ZHANG Y, et al.Application of improved CNN in solar panel defect detection[J]. Computer simulation, 2020, 37(3): 458-463, 474.
[6] 田浩, 周强, 贺晨龙. 基于多尺度特征融合的光伏组件缺陷检测[J]. 计算机工程与应用, 2024, 60(3): 340-347.
TIAN H, ZHOU Q, HE C L.Defect detection of photovoltaic modules based on multi-scale feature fusion[J]. Computer engineering and applications, 2024, 60(3): 340-347.
[7] 孙海蓉, 李莉, 周映杰等. 基于特征金字塔融合高分辨率网络的光伏热斑识别[J]. 太阳能学报, 2023, 44(9): 109-116.
SUN H R, LI L, ZHOU Y J et al. Photovoltaic hot spot recognition based on feature pyramid fusion high-resolution network[J]. Acta energiae solaris sinica, 2023, 44(9):109-116.
[8] 孙航. 基于频域特性的光伏组件热斑故障检测方法研究[D]. 西安: 西安理工大学, 2023.
SUN H.Research on hot spot fault detection method of photovoltaic modules based on frequency domain characteristics[D]. Xi'an: Xi'an University of Technology, 2023.
[9] 朱喆, 黄勇, 来春庆等. 改进YOLOv5的轻量化光伏热斑检测算法[J]. 机电工程技术, 2023, 52(8): 16-21+96.
ZHU Z, HUANG Y, LAI C Q, et al. Improved YOLOv5 lightweight photovoltaic hot spot detection algorithm [J]. Mechanical & electrical engineering technology, 2023, 52 (8): 16-21+96.
[10] 殷杏子, 彭宁宁, 詹学燕. 基于持续同调的过滤式特征选择算法[J]. 计算机科学, 2023,50(6): 159-166.
YIN X Z, PENG N N, ZHAN X Y.Filtered feature selection algorithm based on persistent homology[J]. Computer science, 2023, 50(6): 159-166.
[11] FROSINI P, LANDI C.Size theory as a topological tool for computer vision[J]. Pattern recognition and image analysis, 1999, 9(4): 596-603.
[12] CARLSSON G.Topology and data[J]. Bulletin of the American mathematical society, 2009, 46(2): 255-308.
[13] SINGH G, MEMOLI F, CARLSSON G.Topological methods for the analysis of high dimensional data sets and 3D object recognition[C]// Eurographics symposium on Point-Based graphics, Prague, Czech, 2007.
[14] CARLSSON G, ISHKHANOV T, DE SILVA V, et al.On the local behavior of spaces of natural images[J]. International journal of computer vision, 2008, 76(1): 1-12.
[15] 张景亮, 鞠先孟. 持续同调在图像分类和识别中的应用[J]. 应用数学与计算数学学报, 2017, 31(4): 494-508.
ZHANG J L, JU X M.Application of persistent homology to image classification and recognition[J]. Communication on applied mathematics and computation, 2017, 31(4): 494-508.
[16] 李双悦. 拓扑数据分析及其在图像分类中的应用[D]. 济南: 山东大学, 2022.
LI S Y.Topological data analysis and its application in image classification[D]. Jinan: Shandong University, 2022.
[17] OTTER N, PORTER M A, TILLMANN U, et al.A roadmap for the computation of persistent homology[J]. EPJ data science, 2017, 6(1): 17.
[18] 姚慧. 基于卷积神经网络的光伏阵列热斑故障检测方法研究[D]. 北京: 华北电力大学, 2020.
YAO H.Research on fault detection method of photovoltaic array hot spot based on convolutional neural network[D]. Beijing: north China electric power university, 2020.
[19] 孙海蓉, 潘子杰, 晏勇. 基于深度卷积自编码网络的小样本光伏热斑识别与定位[J]. 华北电力大学学报: 自然科学版, 2021, 48(4): 91-98.
SUN H R, PAN Z J, YAN Y.Identification and location of photovoltaic hot spots with small samples based on deep convolutional autoencoder network[J]. Journal of North China Electric Power University(natural science edition), 2021, 48(4): 91-98.