Ebied Mostafa A, Munshi Amr, Alhuzali Shakir A, El-Sotouhy Mohamed M, Shehta Amr I, Elborlsy M S
Electronics Technology Department, Faculty of Technology and Education, Beni-Suef University, Banī Suwayf, Egypt.
Department of Computer and Network Engineering, College of Computing, Umm Al-Qura University, Makkah, Saudi Arabia.
Sci Rep. 2025 Aug 27;15(1):31640. doi: 10.1038/s41598-025-14478-y.
This paper discusses a deep learning approach for detecting defects in photovoltaic (PV) modules using electroluminescence (EL) images. The method addresses key challenges in two practical areas: Creating high-quality EL images to overcome imbalance issues in existing datasets. This is accomplished by employing generative adversarial network (GAN) properties to generate new images. Enhancing training efficiency and performance through a one-cycle policy with optimized learning rate settings, designed to overcome hardware limitations. The research highlights that while automatic defect classification in PV modules is gaining attention as an alternative to visual/manual inspection, the process remains challenging due to the inhomogeneous nature of cell cracks and complex backgrounds in crystalline solar cells. A comparison was made between popular deep learning models (Densenet169, Densenet201, Resnet101, Resnet152, Senet154, Vgg16, and Vgg19) to assess the effectiveness of our approaches on multiple variants of our dataset. We also observe a shift in the phenomenon of moving the threshold in regression estimates because of employing a policy that uses a dynamic threshold instead of a standard threshold (0.5). We have employed two different categorizations that use binary numbers; the first employs four classes (0%, 33%, 67%, and 100%), while the second employs eight classes that are identical to four classes. However, each class has two varieties (monocrystalline and polycrystalline) and a boundary beyond which results will be obtained. Based on the performance results, it was found that the pre-trained Resnet152 model achieved the highest classification accuracy (90.13% for Datasets) of all approaches. Additionally, we have demonstrated that approaches that utilize over-sampling have the greatest performance. These findings emphasize the strength and innovation of our approach, combining advanced data augmentation, adaptive thresholding, and optimized learning strategies. The proposed system not only achieved a peak classification accuracy of 90.13% using ResNet152 but also demonstrated high robustness, reduced training time, and superior generalization across defect types and cell categories. This positions our framework as a scalable and deployment-ready solution for real-world photovoltaic quality inspection systems.
本文讨论了一种使用电致发光（EL）图像检测光伏（PV）模块缺陷的深度学习方法。该方法解决了两个实际领域中的关键挑战：创建高质量的EL图像以克服现有数据集中的不平衡问题。这是通过利用生成对抗网络（GAN）属性生成新图像来实现的。通过具有优化学习率设置的单周期策略提高训练效率和性能，该策略旨在克服硬件限制。研究强调，虽然光伏模块中的自动缺陷分类作为视觉/手动检查的替代方法正受到关注，但由于晶体太阳能电池中电池裂纹的不均匀性质和复杂背景，该过程仍然具有挑战性。对流行的深度学习模型（Densenet169、Densenet201、Resnet101、Resnet152、Senet154、Vgg16和Vgg19）进行了比较，以评估我们的方法在数据集的多个变体上的有效性。我们还观察到，由于采用了使用动态阈值而不是标准阈值（0.5）的策略，回归估计中移动阈值的现象发生了变化。我们采用了两种使用二进制数字的不同分类；第一种采用四类（0%、33%、67%和100%），而第二种采用与四类相同的八类。然而，每个类别有两个变体（单晶和多晶）以及一个边界，超过该边界将获得结果。根据性能结果，发现预训练的Resnet152模型在所有方法中实现了最高的分类准确率（数据集为90.13%）。此外，我们已经证明，利用过采样的方法具有最大的性能。这些发现强调了我们方法的优势和创新性，结合了先进的数据增强、自适应阈值处理和优化的学习策略。所提出的系统不仅使用ResNet152实现了90.13%的峰值分类准确率，还展示了高鲁棒性、减少的训练时间以及在缺陷类型和电池类别上的卓越泛化能力。这使我们的框架成为适用于实际光伏质量检测系统的可扩展且可部署的解决方案。
