The estimation of series resistance in photovoltaic (PV) cells is a crucial parameter that significantly influences their efficiency and overall performance. This study proposes a novel methodology to predict the slope of the current–voltage (Ⅰ–Ⅴ) curve of a PV cell in the first quadrant, where this slope (the electrical conductance) is directly associated with the series resistance of the cell. By leveraging artificial intelligence techniques, a convolutional neural network model has been developed to estimate this slope from electroluminescence (EL) images of the cells. The model was trained on a dataset consisting of EL images of PV cells with artificially induced defects, together with the corresponding slope values derived from the cells' Ⅰ–Ⅴ curves. Furthermore, this work presents a second model that combines the slope information and EL images to improve the prediction of the maximum power point (MPP) of a PV cell, surpassing previous approaches that rely solely on EL imagery. Both models demonstrated low error rates across multiple evaluation metrics, evidencing their accuracy and robustness. Additionally, comparative analysis with other machine learning methods highlights the competitive performance of the proposed approaches. These contributions provide promising tools for enhancing the assessment and diagnosis of PV cell efficiency and reliability, potentially leading to improved performance and increased longevity of photovoltaic systems.
Citation: Hector Felipe Mateo-Romero, Jose Ignacio Morales Aragonés, Luis Hernández-Callejo, Miguel Ángel González-Rebollo, Valentín Cardeñoso-Payo, Victor Alonso-Gómez, Mario Carbonó delaRosa, Ginés García Mateos. CNN-based estimation of series resistance in photovoltaic cells from electroluminescence images with application to output power prediction[J]. Mathematical Biosciences and Engineering, 2026, 23(5): 1269-1288. doi: 10.3934/mbe.2026046
The estimation of series resistance in photovoltaic (PV) cells is a crucial parameter that significantly influences their efficiency and overall performance. This study proposes a novel methodology to predict the slope of the current–voltage (Ⅰ–Ⅴ) curve of a PV cell in the first quadrant, where this slope (the electrical conductance) is directly associated with the series resistance of the cell. By leveraging artificial intelligence techniques, a convolutional neural network model has been developed to estimate this slope from electroluminescence (EL) images of the cells. The model was trained on a dataset consisting of EL images of PV cells with artificially induced defects, together with the corresponding slope values derived from the cells' Ⅰ–Ⅴ curves. Furthermore, this work presents a second model that combines the slope information and EL images to improve the prediction of the maximum power point (MPP) of a PV cell, surpassing previous approaches that rely solely on EL imagery. Both models demonstrated low error rates across multiple evaluation metrics, evidencing their accuracy and robustness. Additionally, comparative analysis with other machine learning methods highlights the competitive performance of the proposed approaches. These contributions provide promising tools for enhancing the assessment and diagnosis of PV cell efficiency and reliability, potentially leading to improved performance and increased longevity of photovoltaic systems.
| [1] | S. M. Sze, K. K. Ng, Physics of Semiconductor Devices, Wiley, Hoboken, NJ, 3rd edition, 2007. |
| [2] | M. A. Green, Solar Cells: Operating Principles, Technology, and System Applications. Prentice-Hall, Englewood Cliffs, NJ, 1982. |
| [3] |
A. Jain, A. Kapoor, Exact analytical solutions of the parameters of real solar cells using lambert w-function, Solar Energy Mater. Solar Cells, 81 (2004), 269–277. https://doi.org/10.1016/j.solmat.2003.11.018 doi: 10.1016/j.solmat.2003.11.018
|
| [4] |
D. S. H. Chan, J. C. H. Phang, Analytical methods for the extraction of solar-cell single- and double-diode model parameters from i-v characteristicsm, IEEE Transact. Electron. Devices, 34 (1987), 286–293. https://doi.org/10.1109/T-ED.1987.22920 doi: 10.1109/T-ED.1987.22920
|
| [5] |
Y. LeCun, Y. Bengio, G. Hinton, Deep learning, Nature, 521 (2015), 436–444. https://doi.org/10.1038/nature14539 doi: 10.1038/nature14539
|
| [6] | M. A Green, Solar Cells: Operating Principles, Technology, and System Applications. Prentice-Hall, 1982. |
| [7] |
M. Wolf, H. S. Rauschenbach, Series resistance effects on solar cell measurements, Adv. Energy Convers., 3 (1963), 455–479. https://doi.org/10.1016/0365-1789(63)90063-8 doi: 10.1016/0365-1789(63)90063-8
|
| [8] | Y. Hishikawa, T. Kinoshita, K. Tanaka, T. Nishiwaki, Series resistance determination from current-voltage characteristics at different illumination intensities, Progress Photovolt. Res. Appl., 16 (2005), 465–475. |
| [9] |
J. Lee, J. Park, K. Yoon, Machine learning for solar cell design and characterization, J. Power Sources, 451 (2020), 227743. https://doi.org/10.1016/j.jpowsour.2020.227743 doi: 10.1016/j.jpowsour.2020.227743
|
| [10] | Z. Chen, Y. Liu, Y. Yang, Artificial intelligence-based framework for photovoltaic system fault diagnosis and performance prediction, Renew. Sustain. Energy Rev., 136 (2021), 110422. |
| [11] | P. Mehta, R. Ramprasad, Applying machine learning to accelerate material discovery for photovoltaic applications, J. Appl. Phys., 127 (2020), 190901. |
| [12] | W.He, L.Zhang, M. Chen, Data-driven methods for fault detection in photovoltaic arrays, Solar Energy, 193 (2019), 329–343. |
| [13] | K. Patel, S. K. Jha, R.Rajput, Neural network-based approach for PV system performance prediction, Renew. Energy, 172 (2021), 93–104. |
| [14] |
S. Nara, T. Fuyuki, et al. Quantitative prediction of power loss for damaged photovoltaic modules using electroluminescence, Energies, 11 (2018), 1172. https://doi.org/10.3390/en11051172 doi: 10.3390/en11051172
|
| [15] | T. Sauer, M. Rinio, U. Eitner, J. Bagdahn, Characterization of series resistance in photovoltaic modules by combined electroluminescence and infrared thermography imaging, Solar Energy Mater Solar Cell, 95 (2011), 2131–2135. |
| [16] |
O. Breitenstein, M. Langenkamp, Series resistance imaging of solar cells by voltage dependent electroluminescence, Progress Photovolt. Res. Appl., 11 (2003), 515–526. https://doi.org/10.1002/pip.520 doi: 10.1002/pip.520
|
| [17] | T. Fuyuki, H. Kondo, M. Fukawa, A. Kitiyanan, Y. Yamazaki, Quantitative evaluation of electroluminescence images of solar cells, Japanese J. Appl. Phys., 45 (2006), L845. |
| [18] |
H. F. Mateo Romero, M. A. Gonzalez Rebollo, V. Cardenoso-Payo, V. Alonso Gomez, A. Redondo Plaza, R. T. Moyo, et al., Applications of artificial intelligence to photovoltaic systems: A review, Appl. Sci., 12 (2022). https://doi.org/10.3390/app121910056 doi: 10.3390/app121910056
|
| [19] |
A. Mellit, S. A. Kalogirou, Artificial intelligence techniques for photovoltaic applications: A review, Progress Energy Combust. Sci., 34 (2008), 574–632. https://doi.org/10.1016/j.pecs.2008.01.001 doi: 10.1016/j.pecs.2008.01.001
|
| [20] | H. F. Mateo-Romero, M. E. Carbonó dela Rosa, L. Hernández-Callejo, M. González-Rebollo, V. Cardeñoso-Payo, V. Alonso-Gómez, et al., Enhancing photovoltaic cell performance evaluation: An adaptive neural fuzzy inference modeling approach, Progress in Artificial Intelligence, 2024, under review. |
| [21] |
H. F. Mateo-Romero, L. Hernandez-Callejo, M. A. G. Rebollo, V. Cardeñoso-Payo, V. A. Gomez, J. I. M. Aragonés, et al., Optimized estimator of the output power of PV cells using el images and i–v curves, Solar Energy, 265 (2021), 112089. https://doi.org/10.1016/j.solener.2023.112089 doi: 10.1016/j.solener.2023.112089
|
| [22] | H. F. Mateo-Romero, Á. Pérez-Romero, L. Hernández-Callejo, S. Gallardo-Saavedra, V. Alonso-Gómez, J. Morales-Aragonés, et al., Photovoltaic cells defects classification by means of artificial intelligence and electroluminescence images, In: Sergio Nesmachnow and Luis Hernández Callejo, editors, Smart Cities, pages 31–41, Cham, 2022. Springer International Publishing. https://doi.org/10.1007/978-3-030-96753-6_3 |
| [23] |
H. F. Mateo-Romero, L. Hernandez-Callejo, M. A. G. Rebollo, V. Cardeñoso-Payo, V. A. Gomez, H. J. Bello, et al., Synthetic dataset of electroluminescence images of photovoltaic cells by deep convolutional generative adversarial networks, Sustainability, 15 (2023), 7175. https://doi.org/10.3390/su15097175 doi: 10.3390/su15097175
|
| [24] | H. F. Mateo-Romero, J. I. M. Aragonés, L. Hernández-Callejo, M. Á. González-Rebollo, V. Cardeñoso-Payo, V. Alonso-Gómez, et al., Novel convolutional neural network model for estimating series resistance in PV cells by predicting i-v curve slopes on electroluminescence images, In: Sergio Nesmachnow and Luis Hernández Callejo, editors, Smart Cities, pages 105–117, Cham, 2025. Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-85324-1_8 |
| [25] | H. F. Mateo-Romero, J. I. Morales, L. Hernández Callejo, V. Cardeñoso-Payo, M. Gonzalez, V. Alonso Gómez, et al., Dataset of paper "cnn-based estimation of series resistance in photovoltaic cells from electroluminescence images with application to output power prediction", September 2025. https://doi.org/10.5281/zenodo.17083852 |
| [26] |
J. I. Morales-Aragonés, V. A. Gómez, S. Gallardo-Saavedra, A. Redondo-Plaza, D. Fernández-Martínez, L. Hernández-Callejo, Low-cost three-quadrant single solar cell i-v tracer, Appl. Sci., 12 (2022). https://doi.org/10.3390/app12136623 doi: 10.3390/app12136623
|
| [27] | H. F. Mateo-Romero, Employing artificial intelligence techniques for the estimation of energy production in photovoltaic solar cells based on electroluminescence images, PhD thesis, Universidad de Valladolid, 2024. |
| [28] | K. He, X. Zhang, S. Ren, J. Sun, Deep residual learning for image recognition, In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 770–778, 2016. https://doi.org/10.1109/CVPR.2016.90 |
| [29] | D. A. Bonnell, R. J. Nemanich, Characterization of Electronic Materials and Devices, Wiley-VCH, Weinheim, Germany, 2000. |
| [30] | T. O'Malley, E. Bursztein, J. Long, F. Chollet, H. Jin, L. Invernizzi, et al., Keras Tuner, 2019. Available from: https://github.com/keras-team/keras-tuner |
| [31] |
R. Garnett, Bayesian Optimization, Cambridge University Press, 2022. https://doi.org/10.1017/9781108348973 doi: 10.1017/9781108348973
|
| [32] |
C. Shorten, T. M. Khoshgoftaar, A survey on image data augmentation for deep learning, J. Big Data, 6 (2019), 60. https://doi.org/10.1186/s40537-019-0197-0 doi: 10.1186/s40537-019-0197-0
|
Figures(8) / Tables(7)
Hector Felipe Mateo-Romero, Jose Ignacio Morales Aragonés, Luis Hernández-Callejo, Miguel Ángel González-Rebollo, Valentín Cardeñoso-Payo, Victor Alonso-Gómez, Mario Carbonó delaRosa, Ginés García Mateos. CNN-based estimation of series resistance in photovoltaic cells from electroluminescence images with application to output power prediction[J]. Mathematical Biosciences and Engineering, 2026, 23(5): 1269-1288. doi: 10.3934/mbe.2026046
DownLoad: