A Hybrid Gradient Boosting – XGBoost MPPT Strategy for Photovoltaic Systems Under Partial Shading
Rida AMINE, Noureddine EL BARBRI, Hatim AMEZIANE
Abstract. Maximum Power Point Tracking is a key feature in the power conversion of Photovoltaic PV-based power systems, which enables the system to deliver maximum energy based on fluctuating environmental conditions. Conventional MPPT approaches, such as the Perturb and Observe method (P&O) and Incremental Conductance algorithm (INC), have been observed to degrade as a result of multiple local maxima in the power-voltage (P-V) curve during partial shading conditions (PSC). A novel hybrid MPPT technique based on Gradient Boosting Regression and Extreme Gradient Boosting is presented in this paper. The key idea is to exploit the strong robustness of GBR in mid-high (sunny) irradiance and the generalization capability of XGBoost at low (cloudy) irradiance, by leveraging a hysteresis-based switching scheme that takes into account the current-irradiance level and switches between models. Irradiance level from 200 to 1000 W/m² data simulation results demonstrate that the proposed hybrid controller maintains the load power very near to the theoretical maximum, obtaining instantaneous MPPT efficiencies from 95.73% up to 97.37% and global efficiency of 96.83% across the entire profile, which are largely improved over single-model GBR (83.8%) and XGBoost (78.7%) baselines referenced works.
Keywords
Photovoltaic Systems, Maximum Power Point Tracking, Partial Shading, Gradient Boosting, Xgboost, Machine Learning
Published online 4/25/2026, 9 pages
Copyright © 2026 by the author(s)
Published under license by Materials Research Forum LLC., Millersville PA, USA
Citation: Rida AMINE, Noureddine EL BARBRI, Hatim AMEZIANE, A Hybrid Gradient Boosting – XGBoost MPPT Strategy for Photovoltaic Systems Under Partial Shading, Materials Research Proceedings, Vol. 64, pp 535-543, 2026
DOI: https://doi.org/10.21741/9781644904091-67
The article was published as article 67 of the book Energy Futures
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 license. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
References
[1] Esram, T., & Chapman, P. L. (2007). Comparison of photovoltaic array maximum power point tracking techniques. IEEE Transactions on energy conversion, 22(2), 439-449. https://doi.org/10.1109/TEC.2006.874230
[2] Femia, Nicola, et al. “Optimization of perturb and observe maximum power point tracking method.” IEEE transactions on power electronics 20.4 (2005): 963-973. https://doi.org/10.1109/TPEL.2005.850975
[3] Hohm, D. P., and M. E_ Ropp. “Comparative study of maximum power point tracking algorithms.” Progress in photovoltaics: Research and Applications 11.1 (2003): 47-62. https://doi.org/10.1002/pip.459
[4] Ishaque, Kashif, and Zainal Salam. “A review of maximum power point tracking techniques of PV system for uniform insolation and partial shading condition.” Renewable and Sustainable Energy Reviews 19 (2013): 475-488. https://doi.org/10.1016/j.rser.2012.11.032
[5] Friedman, Jerome H. “Greedy function approximation: a gradient boosting machine.” Annals of statistics (2001): 1189-1232. https://doi.org/10.1214/aos/1013203451
[6] Chen, Tianqi. “XGBoost: A Scalable Tree Boosting System.” Cornell University (2016). https://doi.org/10.1145/2939672.2939785
[7] Ahmed, Jubaer, and Zainal Salam. “A critical evaluation on maximum power point tracking methods for partial shading in PV systems.” Renewable and Sustainable Energy Reviews 47 (2015): 933-953. https://doi.org/10.1016/j.rser.2015.03.080
[8] Goodfellow, Ian, et al. Deep learning. Vol. 1. No. 2. Cambridge: MIT press, 2016.
[9] R. Amine, N. El Barbri and H. Ameziane, “Optimizing Photovoltaic Performance: A Comparative Analysis of XGBoost and Gradient Boosting for MPPT,” 2025 International Conference on Electrical Systems & Automation (ICESA), Troyes, France, 2025, pp. 1-6, doi: 10.1109/ICESA66763.2025.11280867. https://doi.org/10.1109/ICESA66763.2025.11280867
[10] R. Amine, N. El Barbri and H. Ameziane, “Stacked Ensemble Machine Learning for MPPT Under Partial Shading: A High-Accuracy Data-Driven Approach,” 2025 International Conference on Electrical Systems & Automation (ICESA), Troyes, France, 2025, pp. 1-6, doi: 10.1109/ICESA66763.2025.11280669. https://doi.org/10.1109/ICESA66763.2025.11280669
[11] Amine, R., & Ameziane, H. (2025, June). An Enhanced MPPT Algorithm Based on Adaptive Linear Regression with Contextual Weight Memory (ALR-CWM) for PV Systems. In 2025 17th International Conference on Electronics, Computers and Artificial Intelligence (ECAI) (pp. 1-4). IEEE. https://doi.org/10.1109/ECAI65401.2025.11095559
