{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,16]],"date-time":"2026-09-16T14:17:24Z","timestamp":1789568244788,"version":"build-2803163510"},"reference-count":30,"publisher":"MDPI AG","issue":"14","license":[{"start":{"date-parts":[[2025,7,18]],"date-time":"2025-07-18T00:00:00Z","timestamp":1752796800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"National Key Lab of Aerospace Power System and Plasma Technology, Air Force Engineering University","award":["APSPT202304001"],"award-info":[{"award-number":["APSPT202304001"]}]},{"name":"National Key Lab of Aerospace Power System and Plasma Technology, Air Force Engineering University","award":["24GXFW0093"],"award-info":[{"award-number":["24GXFW0093"]}]},{"name":"Xi\u2019an Science and Technology Plan Project","award":["APSPT202304001"],"award-info":[{"award-number":["APSPT202304001"]}]},{"name":"Xi\u2019an Science and Technology Plan Project","award":["24GXFW0093"],"award-info":[{"award-number":["24GXFW0093"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>The increasing deployment of photovoltaic (PV) systems necessitates robust fault detection mechanisms to ensure operational reliability and safety. Conventional approaches, however, struggle in complex industrial environments characterized by high noise, data incompleteness, and class imbalance. This study proposes an innovative Advanced CNN-BiLSTM architecture integrating multi-scale feature extraction with hierarchical attention to enhance PV fault detection. The proposed framework employs four parallel CNN branches with kernel sizes of 3, 7, 15, and 31 to capture temporal patterns across various time scales. These features are then integrated by an adaptive feature fusion network that utilizes multi-head attention. A two-layer bidirectional LSTM with temporal attention mechanism processes the fused features for final classification. Comprehensive evaluation on the GPVS-Faults dataset using a progressive difficulty validation framework demonstrates exceptional performance improvements. Under extreme industrial conditions, the proposed method achieves 83.25% accuracy, representing a substantial 119.48% relative improvement over baseline CNN-BiLSTM (37.93%). Ablation studies reveal that the multi-scale CNN contributes 28.0% of the total performance improvement, while adaptive feature fusion accounts for 22.0%. Furthermore, the proposed method demonstrates superior robustness under severe noise (\u03c3 = 0.20), high levels of missing data (15%), and significant outlier contamination (8%). These characteristics make the architecture highly suitable for real-world industrial deployment and establish a new paradigm for temporal feature fusion in renewable energy fault detection.<\/jats:p>","DOI":"10.3390\/s25144474","type":"journal-article","created":{"date-parts":[[2025,7,18]],"date-time":"2025-07-18T11:12:34Z","timestamp":1752837154000},"page":"4474","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":9,"title":["Advanced Multi-Scale CNN-BiLSTM for Robust Photovoltaic Fault Detection"],"prefix":"10.3390","volume":"25","author":[{"given":"Xiaojuan","family":"Zhang","sequence":"first","affiliation":[{"name":"College of Aeronautics Engineering, Air Force Engineering University, Xi\u2019an 710051, China"},{"name":"College of Mechanical Engineering, Xijing University, Xi\u2019an 710123, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bo","family":"Jing","sequence":"additional","affiliation":[{"name":"College of Aeronautics Engineering, Air Force Engineering University, Xi\u2019an 710051, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaoxuan","family":"Jiao","sequence":"additional","affiliation":[{"name":"College of Aeronautics Engineering, Air Force Engineering University, Xi\u2019an 710051, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ruixu","family":"Yao","sequence":"additional","affiliation":[{"name":"College of Artificial Intelligence, Xi\u2019an Jiaotong University, Xi\u2019an 710049, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2025,7,18]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Zhang, Q., Zhai, Z., Mao, M., Wang, S., Sun, S., Mei, D., and Hu, Q. (2024). Control and Intelligent Optimization of a Photovoltaic (PV) Inverter System: A Review. Energies, 17.","DOI":"10.3390\/en17071571"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"836","DOI":"10.1126\/science.aaw1845","article-title":"Terawatt-Scale Photovoltaics: Transform Global Energy","volume":"364","author":"Haegel","year":"2019","journal-title":"Science"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"469","DOI":"10.1016\/j.rser.2016.10.076","article-title":"Review of Distributed Generation with Photovoltaic Grid-Connected Systems in Brazil: Challenges and Prospects","volume":"75","author":"Trigoso","year":"2017","journal-title":"Renew. Sustain. Energy Rev."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"393","DOI":"10.1038\/s41558-025-02276-3","article-title":"Worldwide Rooftop Photovoltaic Electricity Generation May Mitigate Global Warming","volume":"15","author":"Zhang","year":"2025","journal-title":"Nat. Clim. Change"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"24829","DOI":"10.1007\/s00521-023-09041-7","article-title":"Photovoltaic System Fault Detection Techniques: A Review","volume":"35","author":"Moawad","year":"2023","journal-title":"Neural Comput. Appl."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"659","DOI":"10.1093\/ijlct\/ctad018","article-title":"Detection, Location, and Diagnosis of Different Faults in Large Solar PV System\u2014A Review","volume":"18","author":"Venkatakrishnan","year":"2023","journal-title":"Int. J. Low-Carbon Technol."},{"key":"ref_7","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_8","doi-asserted-by":"crossref","unstructured":"Lawani, A.O., Adeleke, T.B., and Osuizugbo, I.C. (2024, January 26\u201328). Solar Photovoltaic Systems: A Review of Risks, Fault Detection, and Mitigation Strategies. Proceedings of the 2024 IEEE 5th International Conference on Electro-Computing Technologies for Humanity (NIGERCON), Ado Ekiti, Nigeria.","DOI":"10.1109\/NIGERCON62786.2024.10927051"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"205","DOI":"10.1016\/j.ijepes.2018.06.051","article-title":"Centralized Emergency Control for Multi-Terminal VSC-Based Shipboard Power Systems","volume":"104","author":"Panasetsky","year":"2019","journal-title":"Int. J. Electr. Power Energy Syst."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"122071","DOI":"10.1016\/j.apenergy.2023.122071","article-title":"Fast Fault Detection Method for Photovoltaic Arrays with Adaptive Deep Multiscale Feature Enhancement","volume":"353 Pt A","author":"Gong","year":"2024","journal-title":"Appl. Energy"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"112011","DOI":"10.1016\/j.solener.2023.112011","article-title":"Research on DC Arc Fault Detection in PV Systems Based on Adjacent Multi-Segment Spectral Similarity and Adaptive Threshold Model","volume":"264","author":"Qian","year":"2023","journal-title":"Sol. Energy"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"5898","DOI":"10.1016\/j.egyr.2022.04.043","article-title":"Methods of Photovoltaic Fault Detection and Classification: A Review","volume":"8","author":"Hong","year":"2022","journal-title":"Energy Rep."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Islam, M., Rashel, M.R., Ahmed, T., Islam, A.K.M.K., and Tlem\u00e7ani, M. (2023). Artificial Intelligence in Photovoltaic Fault Identification and Diagnosis: A Systematic Review. Energies, 16.","DOI":"10.3390\/en16217417"},{"key":"ref_14","first-page":"7383","article-title":"Faults Diagnosis in a Photovoltaic System Based on Multivariate Statistical Analysis","volume":"47","author":"Gnetchejo","year":"2025","journal-title":"Energy Sources Part A Recovery Util. Environ. Eff."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Hichri, A., Hajji, M., Mansouri, M., and Bouzrara, K. (2022, January 6). Deep Learning Based Fault Diagnosis in Grid-Connected Photovoltaic Systems. Proceedings of the 19th IEEE International Multi-Conference on Systems, Signals & Devices, Setif, Algeria.","DOI":"10.1109\/SSD54932.2022.9955779"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"221","DOI":"10.1016\/j.solener.2021.07.037","article-title":"A Fault Diagnosis Method for Photovoltaic Module Current Mismatch Based on Numerical Analysis and Statistics","volume":"225","author":"Zhang","year":"2021","journal-title":"Sol. Energy"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"4005","DOI":"10.1016\/j.egyr.2023.03.033","article-title":"Fault Detection and Diagnosis in Grid-Connected PV Systems Under Irradiance Variations","volume":"9","author":"Hajji","year":"2023","journal-title":"Energy Rep."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Hashunao, S., and Mehta, R.K. (2020, January 29\u201330). Fault Analysis of Solar Photovoltaic System. Proceedings of the 5th International Conference on Renewable Energies for Developing Countries (REDEC), Marrakech, Morocco.","DOI":"10.1109\/REDEC49234.2020.9163847"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Taye, M.M. (2023). Understanding of Machine Learning with Deep Learning: Architectures, Workflow, Applications and Future Directions. Computers, 12.","DOI":"10.3390\/computers12050091"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"118974","DOI":"10.1016\/j.enconman.2024.118974","article-title":"Fault Diagnosis of Photovoltaic Array with Multi-Module Fusion Under Hyperparameter Optimization","volume":"319","author":"Gong","year":"2024","journal-title":"Energy Convers. Manag."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Zyout, I., and Oatawneh, A. (2020, January 16). Detection of PV Solar Panel Surface Defects Using Transfer Learning of the Deep Convolutional Neural Networks. Proceedings of the Advances in Science and Engineering Technology International Conferences (ASET), Dubai, United Arab Emirates.","DOI":"10.1109\/ASET48392.2020.9118384"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Akhter, M.N., Mekhilef, S., Mokhlis, H., Almohaimeed, Z.M., Muhammad, M.A., Khairuddin, A.S.M., Akram, R., and Hussain, M.M. (2022). An Hour-Ahead PV Power Forecasting Method Based on an RNN-LSTM Model for Three Different PV Plants. Energies, 15.","DOI":"10.3390\/en15062243"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"127961","DOI":"10.1016\/j.energy.2023.127961","article-title":"Memory Long and Short Term Time Series Network for Ultra-Short-Term Photovoltaic Power Forecasting","volume":"279","author":"Huang","year":"2023","journal-title":"Energy"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"136593","DOI":"10.1109\/ACCESS.2021.3117004","article-title":"Convergence of Photovoltaic Power Forecasting and Deep Learning: State-of-Art Review","volume":"9","author":"Massaoudi","year":"2021","journal-title":"IEEE Access"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"He, J., and Vashishtha, G. (2024). Construction and Simulation of a Joint Scale Model for Power Electronic Converters Based on Wavelet Decomposition and Reconstruction Algorithms. PLoS ONE, 19.","DOI":"10.1371\/journal.pone.0298590"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"11325","DOI":"10.1007\/s11042-022-12151-4","article-title":"Adaptive Feature Fusion Network Based on Boosted Attention Mechanism for Single Image Dehazing. Multimed","volume":"81","author":"Wang","year":"2022","journal-title":"Tools Appl."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"He, J., Cui, J., Zhang, G., Xue, M., Chu, D., and Zhao, Y. (2022). Spatial\u2013Temporal Seizure Detection with Graph Attention Network and Bi-Directional LSTM Architecture. Biomed. Signal Process. Control, 78.","DOI":"10.1016\/j.bspc.2022.103908"},{"key":"ref_28","unstructured":"Ahsan, S.B., Ikhalas, M., Khan, M.M., Ullah, S., and Zaheer, M.Z. (2025, January 15). ARDGen: Augmentation Regularization for Domain-Generalized Medical Report Generation. Proceedings of the Computer Vision and Pattern Recognition Conference, Vancouver, BC, Canada."},{"key":"ref_29","first-page":"1","article-title":"A Novel Robust Point Cloud Fitting Algorithm Based on Nonlinear Gauss\u2013Helmert Model","volume":"72","author":"Wang","year":"2023","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"106457","DOI":"10.1016\/j.ijepes.2020.106457","article-title":"Real-Time Fault Detection in PV Systems Under MPPT Using PMU and High-Frequency Multi-Sensor Data Through Online PCA-KDE-Based Multivariate KL Divergence","volume":"125","author":"Bakdi","year":"2021","journal-title":"Int. J. Electr. Power Energy Syst."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/25\/14\/4474\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T18:12:06Z","timestamp":1760033526000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/25\/14\/4474"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,18]]},"references-count":30,"journal-issue":{"issue":"14","published-online":{"date-parts":[[2025,7]]}},"alternative-id":["s25144474"],"URL":"https:\/\/doi.org\/10.3390\/s25144474","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,18]]}}}