Aug 2026· Energies· Vol 19, pp. 3959· 0 citations· 37 references
TL;DR
A Hybrid Fuzzy Convolutional Neural Network (HFCNN) that integrates a fuzzy dense layer utilizing Ordered Fuzzy Numbers (OFNs) into the CNN architecture, establishing the feasibility of integrating Ordered Fuzzy Numbers into CNN architectures without requiring expert membership function design.
Abstract
Convolutional Neural Networks (CNNs) are fundamental tools for image analysis and recognition in monitoring systems, particularly in photovoltaic (PV) installations, where visual inspection and thermal imaging play crucial roles in anomaly detection and energy optimization. This publication presents a Hybrid Fuzzy Convolutional Neural Network (HFCNN) that integrates a fuzzy dense layer utilizing Ordered Fuzzy Numbers (OFNs) into the CNN architecture. The architecture is additionally validated on the public ELPV benchmark of 2624 electroluminescence images of photovoltaic cells, where the HFCNN with Mean of Maxima defuzzification attains classification quality statistically indistinguishable from a CNN baseline while using a four times smaller dense layer and training two to three times faster. The methodology combines the feature extraction capabilities of traditional CNNs with the uncertainty handling properties of fuzzy logic. Experiments using the MNIST dataset demonstrate that HFCNN with Mean of Maxima (MOM) defuzzification achieves comparable accuracy to standard CNNs while using significantly fewer parameters (75% reduction in the dense layer). This efficiency gain is advantageous for deployment on edge computing devices. This work constitutes a methodological contribution—establishing, for the first time, the feasibility of integrating Ordered Fuzzy Numbers into CNN architectures without requiring expert membership function design. While the current study validates this approach on MNIST, actual photovoltaic applications require dedicated future research on real PV thermal imagery. Nevertheless, the proposed HFCNN framework could potentially support practical photovoltaic energy system applications in detecting panel degradation, performance anomalies, and autonomous decision-making in large-scale PV installations.
Energy retention from losses is the primary goal of fault detection methodology for photovoltaic (PV) solar systems. A fault detection model should be designed effectively to minimize power and cost waste. We propose a novel fault detection and localization method that leverages deep learning techniques for PV systems. The model is a hybrid semantic segmentation method that combines Convolutional Neural Networks (CNN) and multi-level transformer neural networks. We examine our model on four different segmentation datasets, which have varied characteristics and different capturing conditions, to ensure our model generalization. Using aerial inspection, the first thermal dataset images give a high performance in locating the faulty cells in PV arrays with an accuracy and global precision of 99.89%, a mean average precision (mAP) of 88.73%, a mean intersection over union (mIoU) of 76.29%, and 84.48% of mean dice (mDice, or mean/unweighted F1 score). We use three electroluminescence datasets to investigate the precise location and class of different fault types and minor cracks at the cell level. The second dataset result achieved perfect detection and segmentation of anomaly areas in cells with 99% accuracy, 96.36% mIoU, 98.13% mDice, and 98.41% mAP. The first two datasets contain binary segmentation images with fault and no-fault classes; to accommodate multiple classes, we have utilized the third and fourth datasets. The third dataset, comprising five PV failure classes, is evaluated against other models, yielding a superior performance of 96.27% precision, 77.64% mAP, 57.3% mIoU, and 68.45% mDice. The final dataset has 29 segmentation classes for testing 25 classes, for which it achieves a precision of 95.05%, 66.7% mAP, 49.3% mIoU, and 59.3% mDice.
E. A. Ramadan, Nada M. Moawad, B. Abou-Zalam et al.· Scientific Reports· 0 citations
This paper investigates the use of artificial intelligence techniques for monitoring photovoltaic systems and for fault detection and diagnosis. The proposed methodology uses approximately 1.37 million records collected over 16 days, including electrical variables, such as the voltage and DC current of the two photovoltaic strings, as well as meteorological parameters, such as solar irradiation and module temperature. The analysis process is structured in two stages: a binary detection stage, which differentiates normal from faulty operation, and a multiclass diagnostic stage, which identifies the type of fault: short circuit, degradation, open circuit or shading. Random Forest, LSTM and CNN-LSTM models are implemented and compared, to capitalize on both the nonlinear relationships between variables and the temporal dependencies in sequential data. Random Forest provides interpretability by analyzing the importance of features, LSTM captures the temporal evolution of signals, and the CNN-LSTM hybrid architecture combines automatic local feature extraction with temporal modeling. Evaluation based on accuracy, F1 score, and confusion matrices demonstrates high classification performance, with the CNN-LSTM model standing out for its increased robustness in identifying complex and transient photovoltaic defects.
G. Olteanu, I. Făgărășan, M. Dobrea· International Conference on...· 0 citations
Solar photovoltaic (PV) system operation and maintenance are important for achieving global sustainability objectives. Also, the reliability of PV is one thing that has always prevented it from reaching the full product guarantee due to hotspot, diode degradation, shading, and cracks issues. Traditional inspection methods are time-consuming, expensive, and can contain errors, which justifies the development of automation systems. This paper proposes a deep learning-based framework for anomaly detection using high-resolution RGB images, which overcomes the drawbacks of low-resolution grayscale datasets. To improve robustness, a dataset consisting of 20,000 PV module images with eleven fault categories and normal modules was systematically preprocessed by resizing, augmentation, and quality assurance. Six models, namely, CNN model, AlexNet, VGG16, ResNet18, DenseNet, and EfficientNetV2B0, were compared with each other through the evaluation metrics of accuracy, precision, recall, and F1 score. The experimental results show that the RGB transform can greatly benefit feature learning and model generalization. Overall, ResNet18 had the highest accuracy (91.1%) while EfficientNetV2B0 had balanced performance overall metrics. The results highlight Deep-Learning applications with fine quality datasets (i.e., achieving high performances of Anomaly Detection, Predictive Maintenance, and Sustainable Energy Generation).
Musab Bin Khaleeq, M. Sarfraz· Strategic Planning for Energ...· 0 citations
This study aimed to develop and implement a cloud-enabled intelligent fault inference system for a photovoltaic installation, using locally acquired electrical and environmental data, preprocessed prior to cloud-based inference.
Claudemiro de Lima Júnior, Mariana da S. M. Sobral, Paulo F. C. Barbosa· Revista Brasileira de Engenh...· 0 citations
Research on fault diagnosis of photovoltaic (PV) power generation has long been troubled by two bottlenecks. First, the simple concatenation or fusion of multi‐source heterogeneous data fails to effectively utilize the embedded information. Second, traditional diagnostic models lack the capacity to generalize effectively under complex operating conditions. Therefore, this paper proposes a multi‐source data fusion diagnosis model based on an adaptive fusion mechanism: by constructing a “data layer‐feature layer” collaborative architecture, it can realize the fusion of signals from the same source at the data layer and the dynamic fusion of heterogeneous information at the feature layer by using Adaptive Convolutional Neural Network (Adaptive CNN), leveraging the advantages of spatial feature extraction through signal‐image conversion. Experimental results show that the accuracy of the model reaches 99.0% on the dataset, and the macro‐average F1‐Score is improved to 98.8%. Compared with the traditional direct splicing and fusion methods, the accuracy and F1‐Score are improved by 10.3% and 11.7%, respectively. At the same time, this study not only verifies the advantages of the adaptive hybrid fusion framework but also provides a technical route with high engineering application value for high‐reliability PV intelligent operation and maintenance systems.
Zheng Li, Lin Wang, Huanghuang Jin et al.· Engineering Reports· 0 citations
