Power, Control, and Data Processing Systems

Fault Detection and Classification in PV Arrays with Machine Learning Algorithms

Document Type : Original Research

Authors

Afshin Najafi
Mohammadhossein Tamimi
Ehsan Shahi

Faculty of Electrical Engineering, K. N. Toosi University of Technology, Tehran, Iran

https://doi.org/10.30511/pcdp.2026.2092605.1089
Abstract
Fault detection and classification in photovoltaic arrays are important for

improving the reliability, safety, and energy production of photovoltaic systems.

This study presents a two-stage evaluation of machine learning-based PV fault

diagnosis using both simulation-generated and measured experimental data. First,

a 64-module PV array with an 8×8 configuration is modeled in

MATLAB/Simulink under ten normal and faulty operating conditions. Support

Vector Machine and Naive Bayes classifiers are evaluated before and after

Principal Component Analysis preprocessing. PCA-SVM achieves the highest

accuracy of 97.93% in the ten-class simulation study, compared with 86.50% and

43.125% for SVM and Naive Bayes, respectively. Second, external validation is

performed using twelve measured current-voltage curves obtained from six

physical PV panels under clean and partial-shading conditions. The measured

curves are processed into 3,600 observations and evaluated using leave-one-panel

out validation. Six classifier families, including SVM, Naive Bayes, Random

Forest, XGBoost, LightGBM, and MLP are investigated both with and without

PCA. Random Forest achieves the highest experimental-data accuracy of 97.53%.

PCA increases SVM accuracy from 68.14% to 79.69% and XGBoost accuracy

from 83.33% to 90.03%, while its effect is unfavorable for several other classifiers.

The results demonstrate that PCA can improve feature separability, but its

effectiveness is classifier- and dataset-dependent. The external validation confirms

the usefulness of electrical and environmental features for measured partial

shading diagnosis while also highlighting the need for broader field validation.

Keywords

Photovoltaic array
Fault diagnosis
Fault classification
Principal component analysis
External validation
Support vector machine
Random forest
Subjects
Reference
[1] International Renewable Energy Agency. Renewable Capacity Statistics 2026. Abu Dhabi: IRENA; 2026. Available: https://www.irena.org//media/Files/IRENA/Agency/Publication/2026/ Mar/IRENA_DAT_RE_capacity_statistics_2026.pdf.
[2] Rahman MM and Srivastava AK. Detection, classification, and localization of faults and failures in photovoltaic arrays: a review, comparative analysis, and path forward. Discover Energy 2026;6:26. doi: 10.1007/s43937-026-00136-5.
[3] Madeti SR and Singh SN. A comprehensive study on different types of faults and detection techniques for solar photovoltaic system. Solar Energy 2017;158:161–185. doi: 10.1016/j.solener.2017.08.069.
[4] Pillai DS and Rajasekar N. A comprehensive review on protection challenges and fault diagnosis in PV systems. Renewable and Sustainable Energy Reviews 2018;91:18–40. doi: 10.1016/j.rser.2018.03.082.
[5] Zbib B and Al Sheikh H. Fault detection and diagnosis of photovoltaic systems through I–V curve analysis. 2nd International Conference on Electrical, Communication, and Computer Engineering 2020:1–6. doi: 10.1109/ICECCE49384.2020.9179390.
[6] Madeti SR and Singh SN. Modeling of PV system based on experimental data for fault detection using kNN method. Solar Energy 2018;173:139–151. doi: 10.1016/j.solener.2018.07.038.
[7] Chen Z et al. Random forest based intelligent fault diagnosis for PV arrays using array voltage and string currents. Energy Conversion and Management 2018;178:250–264. doi: 10.1016/j.enconman.2018.10.040.
[8] Fadhel S et al. PV shading fault detection and classification based on I–V curve using principal component analysis: Application to isolated PV system. Solar Energy 2019;179:1–10. doi: 10.1016/j.solener.2018.12.048.
[9] Chen LC et al. Fault diagnosis and classification for photovoltaic arrays based on principal component analysis and support vector machine. IOP Conference Series: Earth and Environmental Science 2018;188:012089. doi: 10.1088/1755-1315/188/1/012089.
[10] Yi Z and Etemadi AH. A novel detection algorithm for line-to-line faults in photovoltaic arrays based on support vector machine. IEEE PES General Meeting 2016:1–5. doi: 10.1109/PESGM.2016.7742026.
[11] Vahidi A and Golkar MA. Fault detection and classification in PV arrays using machine learning algorithms in the presence of noisy data. 9th Iranian Conference on Renewable Energy and Distributed Generation 2022:1–7. doi: 10.1109/ICREDG54199.2022.9804510.
[12] Nedaei A, Eskandari A and Aghaei M. Photovoltaic fault detection and classification: Reconsideration of classic machine learning and dataset shrinkage techniques for simplification. Results in Engineering 2025;27:106356. doi: 10.1016/j.rineng.2025.106356.
[13] Hassan MS, Chin VJ and Gopal L. Accurate diagnosis of concurrent faults in photovoltaic systems using CONMI-based feature selection and support vector machines. Energy Conversion and Management 2025;344:120293. doi: 10.1016/j.enconman.2025.120293.
[14] Pillai PK, Devaraj D and Selvam S. Smart diagnostics of AI-powered IoT solutions for solar grid reliability. Discover Electronics 2025;2:72. doi: 10.1007/s44291-025-00113-7.
[15] Ali YM and Ding L. Fault diagnosis in photovoltaic arrays: A robust and efficient approach using feature engineering and 1D-CNN. International Journal of Electrical Power & Energy Systems 2025;170:110863. doi: 10.1016/j.ijepes.2025.110863.
[16] Sabati A, Bayindir R and Rahebi J. Photovoltaic panels fault detection with convolutional neural network and Bitterling Fish Optimization algorithm. International Journal of Computational Intelligence Systems 2025;18:239. doi: 10.1007/s44196-025-00984-4.
[17] Navid Q, Xia Y, Ali A, Hassan A, Ali A, Manzoor T and Ali S. Fault diagnosis framework for photovoltaic system using fully connected neural network-based back surface thermal analysis. Solar Energy 2026;303:114139. doi: 10.1016/j.solener.2025.114139.
[18] Seghiour A et al. Fault detection and diagnosis in photovoltaic systems using artificial intelligence and time-frequency analysis. Scientific Reports 2026;16:10056. doi: 10.1038/s41598-026-39386-7.
[19] Talaat FM, Salem M and Shaban WM. AI-driven fault detection and classification in photovoltaic systems using deep learning techniques. Scientific Reports 2026;16:8727. doi: 10.1038/s41598-026-40246-7.
[20] Vahidi A, Toulabi M, Pourmoosavi MA and Amraee T. Semi-supervised fault detection and classification in photovoltaic arrays using a modified generative adversarial networks architecture. Energy Reports 2026;16:109516. doi: 10.1016/j.egyr.2026.109516.
[21] Vahidi A, Taghizadeh A and Toulabi M. Fault detection and classification in induction motors: An explainable convolutional neural networks approach. 2025 5th International Conference on Electrical Machines and Drives 2025. doi: 10.1109/ICEMD68435.2025.11441724.
[22] Vahidi A, Nikoofard A and Toulabi M. Data-driven incomplete information Stackelberg game approach for economic load dispatch in power grid. Electric Power Systems Research 2026;259:113024. doi: 10.1016/j.epsr.2026.113024.
[23] Restrepo-Cuestas BJ, Guarnizo-Lemus C, Montoya-Marín JA and Montano J. Dataset of photovoltaic panel performance under different fault conditions: cracks, discoloration, and shading effects. Data in Brief 2025;59:111392. doi: 10.1016/j.dib.2025.111392.
[24] Restrepo-Cuestas B, Guarnizo-Lemus C, Rojas-Montano JJ, MontoyaMarin JA and Ibañez-Diaz AA. Experimental data for mismatching faults of photovoltaic modules. Mendeley Data, Version 2; 2024. doi: 10.17632/xjs42j8dtf.2.
Power, Control, and Data Processing Systems
Volume 3, Issue 4
Autumn 2026
Pages 67-78

  • Receive Date 26 June 2026
  • Revise Date 14 July 2026
  • Accept Date 21 July 2026
  • First Publish Date 21 July 2026
  • Publish Date 01 December 2026

Home

AI Policy

Submit Manuscript

Contact Us