{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,25]],"date-time":"2026-02-25T15:27:16Z","timestamp":1772033236611,"version":"3.50.1"},"reference-count":34,"publisher":"MDPI AG","issue":"3","license":[{"start":{"date-parts":[[2026,2,25]],"date-time":"2026-02-25T00:00:00Z","timestamp":1771977600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Algorithms"],"abstract":"<jats:p>Photovoltaic (PV) systems nowadays represent an essential component of renewable energy production. However, undetected faults often compromise their reliability, leading to significant energy losses and high maintenance costs. Existing deep learning approaches for PV fault diagnosis have achieved high accuracy, but they require massive, labeled datasets and high computational resources, which make them unsuitable for real-time applications. This paper proposes a lightweight, self-supervised hybrid learning framework for real-time PV fault diagnosis to address these limitations. First, the dataset is split into training, testing, and validation subsets. Thereafter, weighted class calculation steps are performed to overcome the issue of imbalance in the data. Then, a self-supervised pre-training phase is established to enable the encoder to produce effective internal representations prior to the implementation of a supervised fine-tuning classifier, characterized as a lightweight feed-forward network (Dense\u2013Dropout\u2013Dense Softmax), which will be trained using categorical cross-entropy and fault-type labels. Finally, a supervised fine-tuning stage is employed based on the pre-trained hybrid CNN\u2013transformer encoder to perform PV fault classification. The experimental results indicate that the proposed approach outperforms existing models by achieving an overall accuracy of 99.8%, a recall of 99.6%, and an outstanding specificity of 100%. The confusion matrix demonstrates that classification is excellent on all operating types. Runtime analysis indicates that the model processes each sample in 2.78 ms and requires 0.07 MB to store weights of 19,429 parameters, confirming its suitability for real-time deployment. These findings highlight that using a hybrid CNN\u2013Transformer encoder with self-supervised learning can improve fault detection and classification performance while significantly reducing inference time, making it an effective and efficient solution for intelligent PV system monitoring.<\/jats:p>","DOI":"10.3390\/a19030173","type":"journal-article","created":{"date-parts":[[2026,2,25]],"date-time":"2026-02-25T13:37:37Z","timestamp":1772026657000},"page":"173","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Lightweight Self-Supervised Hybrid Learning for Generalizable and Real-Time Fault Diagnosis in Photovoltaic Systems"],"prefix":"10.3390","volume":"19","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-9434-2914","authenticated-orcid":false,"given":"Ghalia","family":"Nassreddine","sequence":"first","affiliation":[{"name":"Computer and Information Systems Department, Rafik Hariri University, Mechref 2010, Lebanon"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9473-2365","authenticated-orcid":false,"given":"Obada","family":"Al-Khatib","sequence":"additional","affiliation":[{"name":"School of Engineering, University of Wollongong in Dubai, Knowledge Village, Dubai P.O. Box 20183, United Arab Emirates"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6197-1025","authenticated-orcid":false,"family":"Imran","sequence":"additional","affiliation":[{"name":"Interdisciplinary Research Center for Intelligent Manufacturing & Robotics, King Fahd University of Petroleum and Minerals, Dhahran 31261, Saudi Arabia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6748-7368","authenticated-orcid":false,"given":"Mohamad","family":"Nassereddine","sequence":"additional","affiliation":[{"name":"School of Engineering, University of Wollongong in Dubai, Knowledge Village, Dubai P.O. Box 20183, United Arab Emirates"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6377-7448","authenticated-orcid":false,"given":"Ali","family":"Hellany","sequence":"additional","affiliation":[{"name":"School of Engineering, Western Sydney University, Sydney, NSW 2751, Australia"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,2,25]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"66787","DOI":"10.1109\/ACCESS.2025.3559587","article-title":"Enhancing the Efficacy of Short-Term Prediction Models for Solar Photovoltaic Systems: An Influence Examination of Chronological and Meteorological Factors","volume":"13","author":"Nassreddine","year":"2025","journal-title":"IEEE Access"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"18","DOI":"10.1016\/j.rser.2018.03.082","article-title":"A comprehensive review on protection challenges and fault diagnosis in PV systems","volume":"91","author":"Pillai","year":"2018","journal-title":"Renew. Sustain. Energy Rev."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"32672","DOI":"10.1109\/ACCESS.2021.3060800","article-title":"LSTM Recurrent Neural Network Classifier for High Impedance Fault Detection in Solar PV Integrated Power System","volume":"9","author":"Veerasamy","year":"2021","journal-title":"IEEE Access"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"126286","DOI":"10.1109\/ACCESS.2021.3110947","article-title":"Deep Learning-Based Fault Diagnosis of Photovoltaic Systems: A Comprehensive Review and Enhancement Prospects","volume":"9","author":"Mansouri","year":"2021","journal-title":"IEEE Access"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Duranay, Z.B. (2023). Fault Detection in Solar Energy Systems: A Deep Learning Approach. Electronics, 12.","DOI":"10.3390\/electronics12214397"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"118076","DOI":"10.1016\/j.enconman.2024.118076","article-title":"Faults detection and diagnosis of PV systems based on machine learning approach using random forest classifier","volume":"301","author":"Amiri","year":"2024","journal-title":"Energy Convers. Manag."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"13852","DOI":"10.1109\/ACCESS.2022.3140287","article-title":"Hybrid Deep Learning Model for Fault Detection and Classification of Grid-Connected Photovoltaic System","volume":"10","author":"Alrifaey","year":"2022","journal-title":"IEEE Access"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"117248","DOI":"10.1016\/j.eswa.2022.117248","article-title":"Comparing multilayer perceptron and probabilistic neural network for PV systems fault detection","volume":"201","author":"Vieira","year":"2022","journal-title":"Expert Syst. Appl."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"41406","DOI":"10.1109\/ACCESS.2025.3547838","article-title":"Fault Detection in Photovoltaic Systems Using a Machine Learning Approach","volume":"13","author":"Zwirtes","year":"2025","journal-title":"IEEE Access"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Quiles-Cucarella, E., S\u00e1nchez-Roca, P., and Agust\u00ed-Mercader, I. (2025). Performance Optimization of Machine-Learning Algorithms for Fault Detection and Diagnosis in PV Systems. Electronics, 14.","DOI":"10.3390\/electronics14091709"},{"key":"ref_11","first-page":"13843","article-title":"Fault classification and identification through machine learning approaches for a solar PV\u2014Battery based water pumping system","volume":"84","author":"Sowmmiya","year":"2025","journal-title":"Multimed. Tools Appl."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"e115","DOI":"10.1002\/ail2.115","article-title":"Fault Detection and Classification for Photovoltaic Panel System Using Machine Learning Techniques","volume":"6","author":"Nassreddine","year":"2025","journal-title":"Appl. AI Lett."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"4325","DOI":"10.1016\/j.egyr.2023.03.094","article-title":"Fault detection and computation of power in PV cells under faulty conditions using deep-learning","volume":"9","author":"Sohail","year":"2023","journal-title":"Energy Rep."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Liu, X., Goh, H.H., Xie, H., He, T., Yew, W.K., Zhang, D., Dai, W., and Kurniawan, T.A. (2025). ResGRU: A Novel Hybrid Deep Learning Model for Compound Fault Diagnosis in Photovoltaic Arrays Considering Dust Impact. Sensors, 25.","DOI":"10.3390\/s25041035"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"118537","DOI":"10.1016\/j.measurement.2025.118537","article-title":"Deep learning based vision transformer approach for detecting overlapping PV faults using multi labeling","volume":"256","author":"Khan","year":"2025","journal-title":"Measurement"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"2611","DOI":"10.1109\/TSG.2025.3531764","article-title":"S3Former: A Deep Learning Approach to High Resolution Solar PV Profiling","volume":"16","author":"Tran","year":"2025","journal-title":"IEEE Trans. Smart Grid"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Lazzaretti, A.E., de Costa, C.H., Rodrigues, M.P., Yamada, G.D., Lexinoski, G., Moritz, G.L., Oroski, E., de Goes, R.E., Linhares, R.R., and Stadzisz, P.C. (2020). A Monitoring System for Online Fault Detection and Classification in Photovoltaic Plants. Sensors, 20.","DOI":"10.3390\/s20174688"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"3584406","DOI":"10.1155\/2022\/3584406","article-title":"Performance Comparison of New Adjusted Min-Max with Decimal Scaling and Statistical Column Normalization Methods for Artificial Neural Network Classification","volume":"2022","author":"Sinsomboonthong","year":"2022","journal-title":"Int. J. Math. Math. Sci."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"117810","DOI":"10.1016\/j.measurement.2025.117810","article-title":"Data reconstruction leverages one-dimensional Convolutional Neural Networks (1DCNN) combined with Long Short-Term Memory (LSTM) networks for Structural Health Monitoring (SHM)","volume":"253","author":"Minh","year":"2025","journal-title":"Measurement"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"71","DOI":"10.1186\/s40537-025-01119-4","article-title":"Resampling approaches to handle class imbalance: A review from a data perspective","volume":"12","author":"Carvalho","year":"2025","journal-title":"J. Big Data"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"152870","DOI":"10.1109\/ACCESS.2025.3604427","article-title":"A Novel Approach for Mitigating Class Imbalance in Arabic Text Classification","volume":"13","author":"Nabil","year":"2025","journal-title":"IEEE Access"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Rezaei-Dastjerdehei, M.R., Mijani, A., and Fatemizadeh, E. (2020, January 26\u201327). Addressing Imbalance in Multi-Label Classification Using Weighted Cross Entropy Loss Function. Proceedings of the 2020 27th National and 5th International Iranian Conference on Biomedical Engineering (ICBME), Tehran, Iran.","DOI":"10.1109\/ICBME51989.2020.9319440"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"4785","DOI":"10.1007\/s10994-023-06326-9","article-title":"Understanding CNN fragility when learning with imbalanced data","volume":"113","author":"Dablain","year":"2024","journal-title":"Mach. Learn."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Nassreddine, G., Nassereddine, M., and Al-Khatib, O. (2025). Ensemble Learning for Network Intrusion Detection Based on Correlation and Embedded Feature Selection Techniques. Computers, 14.","DOI":"10.3390\/computers14030082"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Abed, A.E., Nassreddine, G., Al-Khatib, O., Nassereddine, M., and Hellany, A. (2025). Explainable and Optuna-Optimized Machine Learning for Battery Thermal Runaway Prediction Under Class Imbalance Conditions. Thermo, 5.","DOI":"10.3390\/thermo5030023"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"114340","DOI":"10.1016\/j.knosys.2025.114340","article-title":"A self-supervised pretraining model for time series classification based on data preprocessing","volume":"329","author":"Zhang","year":"2025","journal-title":"Knowl.-Based Syst."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"121576","DOI":"10.1016\/j.ins.2024.121576","article-title":"TS-MAE: A masked autoencoder for time series representation learning","volume":"690","author":"Liu","year":"2025","journal-title":"Inf. Sci."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"525","DOI":"10.1109\/TPAMI.2023.3322604","article-title":"Self-Supervised Masked Convolutional Transformer Block for Anomaly Detection","volume":"46","author":"Madan","year":"2024","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"185","DOI":"10.1016\/j.compchemeng.2018.04.009","article-title":"Deep convolutional neural network model based chemical process fault diagnosis","volume":"115","author":"Wu","year":"2018","journal-title":"Comput. Chem. Eng."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Zhang, S., Zhou, J., Ma, X., Pirttikangas, S., and Yang, C. (2024). TSViT: A Time Series Vision Transformer for Fault Diagnosis of Rotating Machinery. Appl. Sci., 14.","DOI":"10.3390\/app142310781"},{"key":"ref_31","unstructured":"Awe, O.O., and Vance, E.A. (2025). Machine Learning Evaluation of Imbalanced Health Data: A Comparative Analysis of Balanced Accuracy, MCC, and F1 Score. Practical Statistical Learning and Data Science Methods, Springer Nature. STEAM-H: Science, Technology, Engineering, Agriculture, Mathematics & Health."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"609","DOI":"10.1007\/s13389-024-00361-5","article-title":"Regularizers to the rescue: Fighting overfitting in deep learning-based side-channel analysis","volume":"14","author":"Rezaeezade","year":"2024","journal-title":"J. Cryptogr. Eng."},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Bougoffa, M., Benmoussa, S., Djeziri, M., and Palais, O. (2025). Hybrid Deep Learning for Fault Diagnosis in Photovoltaic Systems. Machines, 13.","DOI":"10.3390\/machines13050378"},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Memon, S.A., Javed, Q., Kim, W.-G., Mahmood, Z., Khan, U., and Shahzad, M. (2022). A Machine-Learning-Based Robust Classification Method for PV Panel Faults. Sensors, 22.","DOI":"10.3390\/s22218515"}],"container-title":["Algorithms"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1999-4893\/19\/3\/173\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,25]],"date-time":"2026-02-25T13:41:29Z","timestamp":1772026889000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1999-4893\/19\/3\/173"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,25]]},"references-count":34,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,3]]}},"alternative-id":["a19030173"],"URL":"https:\/\/doi.org\/10.3390\/a19030173","relation":{},"ISSN":["1999-4893"],"issn-type":[{"value":"1999-4893","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,25]]}}}