基于多尺度CNN的光伏组件缺陷识别

doi:10.19912/j.0254-0096.tynxb.2020-0408

太阳能学报 ›› 2022, Vol. 43 ›› Issue (2): 211-216.DOI: 10.19912/j.0254-0096.tynxb.2020-0408

基于多尺度CNN的光伏组件缺陷识别

周颖^1,2, 叶红¹, 王彤¹, 常明新¹

1.河北工业大学人工智能与数据科学学院,天津 300130;
2.河北省控制工程技术研究中心,天津 300130

收稿日期:2020-05-07 出版日期:2022-02-28 发布日期:2022-08-28
通讯作者: 周颖（1971—）,女,博士、副教授,主要从事智能控制与模式识别方面的研究。zhouying2007@163.com
基金资助:
国家自然科学基金（60741307）

PHOTOVOLTAIC MODULE DEFECT IDENTIFICATION BASED ON MULIT-SCALE CONVOLUTION NEURAL NETWORK

Zhou Ying^1,2, Ye Hong¹, Wang Tong¹, Chang Mingxin¹

1. School of Artificial Intelligence, Hebei University of Technology, Tianjin 300130, China;
2. China Hebei Control Engineering Research Center, Tinajin 300130, China

Received:2020-05-07 Online:2022-02-28 Published:2022-08-28

摘要/Abstract

摘要： 为提高光伏组件缺陷分类精度与效率,提出一种改进的多尺度卷积神经网络模型（IMCNN）。该算法根据光伏组件缺陷特点,构建3个不同尺度端对端的卷积神经网络模型,同时为优化网络结构,在3个通道中均引入SE-Inception模块。首先由多通道卷积提取精密度不同的特征;再将这些特征进行融合,得到特征的增强表达;最后实现光伏组件的缺陷分类。由于光伏组件的缺陷样本较少,使用生成对抗网络生成一部分图像样本,达到有效进行数据增强的目的。实验结果表明,所提算法的Kappa系数较高,分类精度与效率均有明显提升。

关键词: 光伏组件, 缺陷识别, 图像处理, 卷积神经网络, 生成对抗网络

Abstract: In order to improve the accuracy and efficiency of photovoltaic module defect classification, an Improved Multi-scale Convolutional Neural Network （IMCNN） model is proposed. According to the defect characteristics of photovoltaic module, the algorithm constructs the end-to-end Convolutional Neural Network model of three channels with different scales. At the same time, in order to optimize the network structure, SE-Inception modules are introduced in all three channels. Firstly, multi-channel convolution is used to extract features with different fineness. Then these features are fused to obtain the enhanced expression of high-level features. Finally, the defect classification of photovoltaic module is realized. Due to the small number of defective samples of photovoltaic module, a part of the image samples is generated using the Generative Adversarial Networks to effectively expanding the dataset. The results show that the Kappa coefficient of the proposed algorithm is higher, and the identification accuracy and efficiency are significantly improved.

Key words: photovoltaic modules, defect identification, picture processing, convolutional neural networks, generative adversarial networks

中图分类号:

TP391.5

周颖, 叶红, 王彤, 常明新. 基于多尺度CNN的光伏组件缺陷识别[J]. 太阳能学报, 2022, 43(2): 211-216.

Zhou Ying, Ye Hong, Wang Tong, Chang Mingxin. PHOTOVOLTAIC MODULE DEFECT IDENTIFICATION BASED ON MULIT-SCALE CONVOLUTION NEURAL NETWORK[J]. Acta Energiae Solaris Sinica, 2022, 43(2): 211-216.

参考文献

[1] BROOKS W S M, LAMB D A, IRVINE S J C. IR reflectance imaging for crystalline Si solar cell crack detection[J]. IEEE journal of photovoltaics, 2015, 5(5): 1271-1275.
[2] 沈凯, 陈克, 杭丽君, 等. 一种可提升光伏系统在局部阴影时发电效率的阵列连接策略[J]. 太阳能学报, 2020, 41(4): 51-58.
SHEN K, CHEN K, HANG L J, et al.A arry strategy for efficiency improvement of photovoltaic systems under partially shaded condition[J]. Acta energiae solaris sinica, 2020, 41(4): 51-58.
[3] STROMER D, VETTER A, OEZKAN H C, et al.Enhanced crack segmentation(eCS): a reference algorithm for segmenting cracks in multicrystalline silicon solar cells[J]. IEEE journal of photovoltaic, 2019, 9(3): 752-758.
[4] JEN C Y, RICHTER C.Doping profile recognition applied to silicon photovoltaic cells using terahertz time-domain spectroscopy[J]. IEEE transactions on terahertz science and technology, 2014, 4(5): 560-567.
[5] SUN Q M, MELNIKOV A, MANDELIS A.Camera-based high frequency heterodyne lock-in carrierographic (frequency-domain photoluminescence)imaging of crystalline silicon wafers[J]. Physica status solidi, 2016, 213(2): 405-411.
[6] 鲁伟明, 李省, 张付特, 等. 基于不同电压下的电致发光和红外成像的太阳能电池缺陷检测[J]. 发光学报, 2014, 35(12): 1511-1519.
LU W M, LI X, ZHANG F T, et al.Defect detection of solar cell based on electroluminescence and thermography imaging with different bias voltage[J]. Chinese journal of luminescence, 2014, 35(12): 1511-1519.
[7] DEMANT M, WELSCHEHOLD T, OSWALD M, et al.Microcracks in silicon wafers I: inline detection and implications of crack morphology on wafer strength[J]. IEEE journal of photovoltaics, 2016, 6(1): 126-135.
[8] GOODFELLOW I J, POUGET-ABADIE J, MIRZA M, et al.Generative adversarial networks[C]//Proceedings of the 27th International Conference on Neural Information Processing Systems, MIT Press, NIPS, 2014.
[9] RADFORD A, METZ L, CHINTALA S.Unsupervised representation learning with deep convolutional generative adversarial networks[J/OL]. [2015-11-19].https://arxiv.org/abs/1511.06434.
[10] LI Y C, XIAO N F, OUYANG W L.Improved boundary equilibrium generative adversarial networks[J]. IEEE access, 2018, 6: 11342-11348.
[11] ARJOVSKY M, CHINTALA S, BOTTOU L.Wasserstein GAN[J/OL].[2017-12-06].https:// arxiv.org/abs/1701.07875.
[12] CHAWLA N V, BOWYER K W, HALL L O, et al.SMOTE: synthetic minority over-sampling technique[J]. Journal of artificial intelligence research, 2002, 16: 321-357.
[13] INOUE H.Data augmentation by pairing samples for images classification[J/OL]. [2018-04-11].https:// arxiv.org/abs/1801.02929.
[14] ZHANG H, CISSE M, DAUPHIN Y N, et al.Mixup: beyond empirical risk minimization[J/OL]. [2018-04-27].https://arxiv.org/abs/1710.09412.
[15] 姚明海, 李洁, 王宪保. 基于RPCA的太阳能电池片表面缺陷检测[J]. 计算机学报, 2013, 36(9): 1943-1952.
YAO M H, LI J, WANG X B.Solar cells surface defects detection using RPCA method[J]. Journal of computer science, 2013, 36(9): 1943-1952.
[16] AGHAMOHAMMADI A H, PRABUWONO A S, SAHRAN S, et al.Solar cell panel crack detection using particle swarm optimization algorithm[C]// 2011 International Conference on Pattern Analysis and Intelligence Robotics, Kuala Lumpur, Malaysia, 2011.
[17] TSAI D M, WU S C, CHIU W Y.Defect detection in solar modules using ICA basis images[J]. IEEE transactions on industrial informatics, 2013, 9(1): 122-131.

基于多尺度CNN的光伏组件缺陷识别

PHOTOVOLTAIC MODULE DEFECT IDENTIFICATION BASED ON MULIT-SCALE CONVOLUTION NEURAL NETWORK

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	马铭遥, 王海松, 马文婷, 张志祥, 张兴. 基于S-V特性分析的晶硅光伏组件阴影遮障诊断[J]. 太阳能学报, 2022, 43(9): 64-72.
[2]	赵斌, 谭恒, 何锁盈, 曲宏伟, 白珍, 周腊吾. 高原高寒地区光伏组件背板冷却对输出功率影响的实验研究[J]. 太阳能学报, 2022, 43(8): 122-129.
[3]	方洛迪, 郭倩, 卫东, 郭泽阳, 黄鑫. 遮挡状态下光伏组件多峰输出特性的参数计算与分析[J]. 太阳能学报, 2022, 43(7): 115-121.
[4]	陶志勇, 杜福廷, 任晓奎, 林森. 基于T-VGG的太阳电池片缺陷检测[J]. 太阳能学报, 2022, 43(7): 145-151.
[5]	沈振兴, 刘百诚, 庄建宏, 孙旗霞. 三相行波电帘的电场分布及除尘机理研究[J]. 太阳能学报, 2022, 43(7): 152-158.
[6]	廖东进, 方晓敏, 黄志平. 光伏组件辐照度增益研究[J]. 太阳能学报, 2022, 43(5): 113-118.
[7]	田勇, 颜健, 彭佑多, 刘永祥, 聂笃忠, 谢欣旖. 高能流密度下太阳能密集光伏组件射流冲击冷却特性[J]. 太阳能学报, 2022, 43(5): 119-126.
[8]	余萍, 曹洁. 基于图形特征的双输入卷积神经网络风力机轴承剩余寿命预测[J]. 太阳能学报, 2022, 43(5): 343-350.
[9]	谢路耀, 莫晨飞, 陈怡. 基于并联光伏组件外特性的最大功率点跟踪方法[J]. 太阳能学报, 2022, 43(4): 148-153.
[10]	易磊, 谢雨龙, 曾凡炎, 李嘉琪, 黄海, 陈俊. 基于神经网络的改进扰动观察法MPPT控制[J]. 太阳能学报, 2022, 43(4): 198-203.
[11]	王妮妮, 马萍, 张宏立, 王聪. 基于多尺度深度卷积网络特征融合的滚动轴承故障诊断[J]. 太阳能学报, 2022, 43(4): 351-358.
[12]	鄢伟安, 孔文琪, 刘卫东, 熊玉辉, 张金晶. 具有初始退化随机特征的光伏组件可靠性评估[J]. 太阳能学报, 2022, 43(3): 152-157.
[13]	张臻, 王磊, 秦志光, 伍敏燕, 许传佳, 全鹏. 双面光伏组件发电性能影响因素理论与实验研究[J]. 太阳能学报, 2022, 43(3): 171-179.
[14]	倪超, 王聪, 朱婷婷, 过奕任. 基于CNN-Bi-LSTM的太阳辐照度超短期预测[J]. 太阳能学报, 2022, 43(3): 197-202.
[15]	李翠明, 刘爽, 龚俊. 光伏组件清洁移动机器人地形分类方法研究[J]. 太阳能学报, 2022, 43(3): 210-215.