Day-Ahead Solar Power Forecasting Using a Hybrid Model Combining Regression and Physical Model Chain
DOI:
https://doi.org/10.62146/ijecbe.v3i1.108Keywords:
Hybrid method, NWP, Physical model chain, Day-ahead PV power forecast, XGBoostAbstract
Solar power forecasting is essential for integrating PV plants into power grids, ensuring stability and aiding system operators (SOs) in decision-making. However, existing day-ahead models struggle with rapid weather changes, while deep learning models require extensive historical data, making them impractical for new PV plants.
This study proposes a hybrid approach combining the XGBoost algorithm for hourly solar irradiance prediction using Numerical Weather Prediction (NWP) data and a physical model to convert irradiance into power. The XGBoost model is periodically retrained via a sliding window mechanism to adapt to dynamic weather conditions.
A case study using two years of 271 kWp PV data from NIST (US) and historical NWP data from ECMWF ENS for GHI forecasting, alongside ECMWF HRES for power conversion, demonstrated the method’s effectiveness. Using just one week of historical data for initial training, the model achieved an nRMSE of 13.35%–13.53%, nMAE of 6.9%–7.03%, and nMBE of -2.03% to -0.29%.
The proposed approach improves PV forecasting reliability for new plants with limited data, serving as an intermediary solution until sufficient historical data is available for deep learning models.
References
David Feldman et al. US solar photovoltaic system and energy storage cost benchmark (Q1 2020). Tech.
rep. National Renewable Energy Lab.(NREL), Golden, CO (United States), 2021.
Felix Creutzig et al. “Technological innovation enables low cost climate change mitigation”. In:
Energy Research & Social Science 105 (2023), p. 103276.
Dolf Gielen et al. “The role of renewable energy in the global energy transformation”. In: Energy
Strategy Reviews 24 (2019), pp. 38–50.
Dazhi Yang et al. “A review of solar forecasting, its dependence on atmospheric sciences and im-
plications for grid integration: Towards carbon neutrality”. In: Renewable and Sustainable Energy
Reviews 161 (2022), p. 112348.
Martin János Mayer. “Benefits of physical and machine learning hybridization for photovoltaic
power forecasting”. In: Renewable and Sustainable Energy Reviews 168 (2022), p. 112772.
Hugo TC Pedro et al. “Assessment of machine learning techniques for deterministic and proba-
bilistic intra-hour solar forecasts”. In: Renewable Energy 123 (2018), pp. 191–203.
Lennard Visser, Tarek AlSkaif, and Wilfried van Sark. “Operational day-ahead solar power fore-
casting for aggregated PV systems with a varying spatial distribution”. In: Renewable Energy 183
(2022), pp. 267–282.
Kejun Wang, Xiaoxia Qi, and Hongda Liu. “A comparison of day-ahead photovoltaic power fore-
casting models based on deep learning neural network”. In: Applied Energy 251 (2019), p. 113315.
Wenting Wang et al. “An archived dataset from the ECMWF Ensemble Prediction System for
probabilistic solar power forecasting”. In: Solar Energy 248 (2022), pp. 64–75.
Dazhi Yang, Wenting Wang, and Tao Hong. “A historical weather forecast dataset from the Eu-
ropean Centre for Medium-Range Weather Forecasts (ECMWF) for energy forecasting”. In: Solar
Energy 232 (Jan. 2022), pp. 263–274. ISSN: 0038092X. DOI: 10.1016/j.solener.2021.12.011.
Christina Brester et al. “Evaluating neural network models in site-specific solar PV forecasting us-
ing numerical weather prediction data and weather observations”. In: Renewable Energy 207 (2023),
pp. 266–274.
Alejandro Catalina, Carlos M. Alaiz, and Jose R. Dorronsoro. “Combining numerical weather pre-
dictions and satellite data for PV energy nowcasting”. In: IEEE Transactions on Sustainable Energy
3 (July 2020), pp. 1930–1937. ISSN: 1949-3029. DOI: 10.1109/TSTE.2019.2946621.
Martin János Mayer and Gyula Gróf. “Extensive comparison of physical models for photovoltaic
power forecasting”. In: Applied Energy 283 (2021), p. 116239.
Dazhi Yang. “Standard of reference in operational day-ahead deterministic solar forecasting”. In:
Journal of Renewable and Sustainable Energy 11.5 (Sept. 2019), p. 053702. ISSN: 1941-7012. DOI: 10.
/1.5114985. eprint: https://pubs.aip.org/aip/jrse/article- pdf/doi/10.1063/1.5114985/
/053702_1_online.pdf . URL: https://doi.org/10.1063/1.5114985.
Yuri V. Makarov et al. “Incorporating Uncertainty of Wind Power Generation Forecast Into Power
System Operation, Dispatch, and Unit Commitment Procedures”. In: IEEE Transactions on Sustain-
able Energy 2.4 (2011), pp. 433–442. DOI: 10.1109/TSTE.2011.2159254.
Dazhi Yang et al. “Operational solar forecasting for grid integration: Standards, challenges, and
outlook”. In: Solar Energy 224 (2021), pp. 930–937.
Martin János Mayer and Dazhi Yang. “Pairing ensemble numerical weather prediction with en-
semble physical model chain for probabilistic photovoltaic power forecasting”. In: Renewable and
Sustainable Energy Reviews 175 (2023), p. 113171.
Rui Guo et al. “Degradation state recognition of piston pump based on ICEEMDAN and XGBoost”.
In: Applied Sciences 10.18 (2020), p. 6593.
DG Erbs, SA Klein, and JA Duffie. “Estimation of the diffuse radiation fraction for hourly, daily
and monthly-average global radiation”. In: Solar Energy 28.4 (1982), pp. 293–302.
Richard Perez et al. “Modeling daylight availability and irradiance components from direct and
global irradiance”. In: Solar Energy 44.5 (1990), pp. 271–289.
Anton Driesse, Adam R Jensen, and Richard Perez. “A continuous form of the Perez diffuse sky
model for forward and reverse transposition”. In: Solar Energy 267 (Jan. 2024), p. 112093.
AF Souka and HH Safwat. “Determination of the optimum orientations for the double-exposure,
flat-plate collector and its reflectors”. In: Solar Energy 10.4 (1966), pp. 170–174.
Widalys De Soto, Sanford A Klein, and William A Beckman. “Improvement and validation of a
model for photovoltaic array performance”. In: Solar Energy 80.1 (2006), pp. 78–88.
N Martin and JM Ruiz. “Corrigendum to “Calculation of the PV modules angular losses under field
conditions by means of an analytical model” [Solar Energy Materials and Solar Cells 70 (1) (2001)
–38]”. In: Solar Energy Materials and Solar Cells 110 (2013), p. 154.
Jay A Kratochvil, William Earl Boyson, and David L King. Photovoltaic array performance model.
Tech. rep. Sandia National Laboratories (SNL), Albuquerque, NM, and Livermore, CA, 2004.
David Faiman. “Assessing the outdoor operating temperature of photovoltaic modules”. In: Progress
in Photovoltaics: Research and Applications 16.4 (2008), pp. 307–315.
Paul Gilman et al. “SAM photovoltaic model technical reference update”. In: NREL: Golden, CO,
USA (2018).
Aron P Dobos. PVWatts version 5 manual. Tech. rep. National Renewable Energy Lab.(NREL),
Golden, CO (United States), 2014.
DL Evans. “Simplified method for predicting photovoltaic array output”. In: Solar Energy 27.6
(1981), pp. 555–560.
Matthew T Boyd. “High-speed monitoring of multiple grid-connected photovoltaic array config-
urations and supplementary weather station”. In: Journal of Solar Energy Engineering 139.3 (2017),
p. 034502.
Matthew Boyd. NIST Weather Station for Photovoltaic and Building System Research. Technical Note
Gaithersburg, MD: National Institute of Standards and Technology, 2016.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 International Journal of Electrical, Computer, and Biomedical Engineering
This work is licensed under a Creative Commons Attribution 4.0 International License.
