Enhanced SARIMAX and Gradient Boosting Load Forecasting for Photovoltaic Integration and Transformer Expansion Planning in a 132/33-kV Substation
Journal of Robotics and Control (JRC) · 01.05.2026 · 7 (3) · 3784-3801
DOI: 10.18196/jrc.v7i3.28869 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Robotics and Control (JRC), online erschienen am 01.05.2026. Verfasst von Anas Mustafa und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
Accurate load forecasting is critical for secure and efficient power system operation in rapidly growing regions such as the Nineveh Governorate in Iraq. This study proposes an integrated framework that couples data-driven daily load forecasting with network scenario analysis to support multi-year expansion planning for an overloaded 132/33 kV substation. The research contribution is an end-to-end pipeline in which daily demand forecasts from a seasonal ARIMA model with exogenous variables (SARIMAX) are used to drive photovoltaic (PV) and transformer reinforcement studies on a real Iraqi grid. A tailored feature-engineering strategy combining climatic, temporal and demographic variables is employed to train and compare two models: a Gradient Boosting regressor and the SARIMAX time-series model. Using strictly time-ordered data from 2022 to 2024, the latter delivers superior accuracy (R² = 0.999998 with very low MAE and RMSE), and residual diagnostics confirm that remaining errors are negligible relative to monthly peak loads. Forecasts for (2025–2030) are then used to define two PVsyst and ETAP expansion scenarios – 18 MW grid-connected PV plant and the installation of the transformer of 25 MVA. Both options provide significant transformer loading reduction and safe operation, and the PV plant option also ensures lower dependence on the upstream grid and emissions, and the second transformer option provides N+1 redundancy in the event of high demand. The modular structure is transferable to other Networks in Iraq after retraining and local validation as was shown in substation tests in Mosul, and could be expanded to include other planning and operational conditions.
Zitieren
APA
Mustafa, A., & Alyozbaky, O. (2026). Enhanced SARIMAX and Gradient Boosting Load Forecasting for Photovoltaic Integration and Transformer Expansion Planning in a 132/33-kV Substation. Journal of Robotics and Control (JRC), 7(3), 3784-3801. https://doi.org/10.18196/jrc.v7i3.28869
Vancouver
Mustafa A, Alyozbaky O. Enhanced SARIMAX and Gradient Boosting Load Forecasting for Photovoltaic Integration and Transformer Expansion Planning in a 132/33-kV Substation. Journal of Robotics and Control (JRC). 2026;7(3):3784-3801. doi:10.18196/jrc.v7i3.28869
BibTeX
@article{Mustafa2026,
author = {Mustafa, Anas and Alyozbaky, Omar},
title = {Enhanced SARIMAX and Gradient Boosting Load Forecasting for Photovoltaic Integration and Transformer Expansion Planning in a 132/33-kV Substation},
journal = {Journal of Robotics and Control (JRC)},
year = {2026},
volume = {7},
number = {3},
pages = {3784--3801},
doi = {10.18196/jrc.v7i3.28869}
}
EUVIMED