Predicting the Shading of Photovoltaic Systems Using Machine Learning
DOI:
https://doi.org/10.52825/pv-symposium.v2i.2636Keywords:
Operation and Maintenance (O&M), Digital Twin, Origins of ShadowsAbstract
In the operation of photovoltaic power plants, precise knowledge of shading is essential in order to carry out differentiated yield and site analyses and to guarantee reliable monitoring and fault detection. A study on the causes of shading carried out with the help of GPT4-o is presented. Subsequently, an innovative approach for predicting shading using a is introduced. By combining physical and data-driven machine learning, it is possible to efficiently complete incomplete shadow analyses and eliminate erroneous data points of a physical model. The presented method utilizes data from 1380 photovoltaic devices with various shading scenarios to train an autoencoder on PV system shading. The autoencoder enables accurate prediction of shading within a detection time of only a few weeks.
Downloads
References
S. Killinger et al., “On the search for representative characteristics of PV systems: Data collection and analysis of PV system azimuth, tilt, capacity, yield and shading,” Sol. En-ergy, vol. 173, pp. 1087–1106, Oct. 2018, doi: 10.1016/j.solener.2018.08.051.
S. R. Pendem and S. Mikkili, “Modelling and performance assessment of PV array topol-ogies under partial shading conditions to mitigate the mismatching power losses,” Sol. Energy, vol. 160, pp. 303–321, Jan. 2018, doi: 10.1016/j.solener.2017.12.010.
M. Schönau et al., “Reliable and Commercially Viable Detection of String Outages in Photovoltaic Plants,” in 2024 International Conference on Renewable Energies and Smart Technologies (REST), Prishtina, Kosovo (UNMIK): IEEE, Jun. 2024, pp. 1–5. doi: 10.1109/REST59987.2024.10645480.
M. Schönau et al., “String Outages in Photovoltaic Plants (Submitted for publication on February 28, 2025),” Renew. Energ., Feb. 2025.
“Hello GPT-4o.” Accessed: Feb. 06, 2025. [Online]. Available: https://openai.com/index/hello-gpt-4o/
“OpenAI Platform.” Accessed: Feb. 06, 2025. [Online]. Available: https://platform.openai.com
A. Vaswani et al., “Attention Is All You Need,” Aug. 01, 2023, arXiv: arXiv:1706.03762. Accessed: Sep. 09, 2024. [Online]. Available: http://arxiv.org/abs/1706.03762
S. Ghosh, J. N. Roy, and C. Chakraborty, “A model to determine soiling, shading and thermal losses from PV yield data,” Clean Energy, vol. 6, no. 2, pp. 372–391, Apr. 2022, doi: 10.1093/ce/zkac014.
S. Sugumar, “A novel on-time partial shading detection technique for electrical reconfigu-ration in solar PV system,” Sol. Energy, 2021, doi: 10.1016/j.solener.2021.07.069.
S. Fadhel, “PV shading fault detection and classification based on I-V curve using princi-pal component analysis_ Application to isolated PV system,” Sol. Energy, 2019, doi: 10.1016/j.solener.2018.12.048.
M. Schönau, D. Daume, B. Hüttl, and D. Landes, “Improving IV Curve Classification by Machine Learning Methods Using Deep Autoencoders,” 40th Eur. Photovolt. Sol. Energy Conf. Exhib., pp. 020410-001-020410–004, 2023, doi: 10.4229/EUPVSEC2023/4CV.1.53.
A. Schulze, M. Panhuysen, D. Daume, and M. Schönau, “Quantitative Shade Detection for PV Systems Based on Clearsky Data,” in 41th European Photovoltaic Solar Energy Conference and Exhibition, Vienna, Sep. 2024., doi: 10.4229/EUPVSEC2024/4BV.3.29
Spatial distribution of daylight: luminance distributions of various reference skies. in Technical report / International Commission on Illumination, no. 110. Vienna: CIE Central Bureau, 1994.
G. E. Hinton and R. R. Salakhutdinov, “Reducing the Dimensionality of Data with Neural Networks,” Science, vol. 313, no. 5786, Art. no. 5786, Jul. 2006, doi: 10.1126/science.1127647.
P. Vincent, H. Larochelle, I. Lajoie, Y. Bengio, and P.-A. Manzagol, “Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local De-noising Criterion”, doi: 10.5555/1756006.1953039.
Y. Lecun, L. Bottou, Y. Bengio, and P. Haffner, “Gradient-based learning applied to doc-ument recognition,” Proc. IEEE, vol. 86, no. 11, pp. 2278–2324, Nov. 1998, doi: 10.1109/5.726791.
K. Simonyan and A. Zisserman, “Very Deep Convolutional Networks for Large-Scale Image Recognition,” Apr. 10, 2015, arXiv: arXiv:1409.1556. Accessed: Sep. 18, 2024. [Online]. Available: http://arxiv.org/abs/1409.1556
M. Schönau, D. Daume, M. Panhuysen, A. Schulze, Hüttl, Bernd, and D. Landes, “Ver-besserte Clear-Sky-Erkennung durch hybrides Maschinelles Lernen,” presented at the RET.Con, Nordhausen, Feb. 2024. Accessed: Sep. 18, 2024. [Online]. Available: https://www.hs-nordhausen.de/fileadmin/Dateien/Veranstaltungen/RETCon_2024_Tagungsband.pdf
Published
How to Cite
Conference Proceedings Volume
Section
License
Copyright (c) 2025 Maximilian Schönau, Joseph Jachmann, Markus Panhuysen, Alexander Schönau, Darwin Daume, Achim Schulze, Bernd Hüttl, Dieter Landes
This work is licensed under a Creative Commons Attribution 4.0 International License.
Accepted 2025-05-02
Published 2025-08-27
Funding data
-
Bayerische Forschungsstiftung
Grant numbers AZ-1564-22