Sliding Window-Based Deep Learning Approach for Solar Power Forecasting in Malaysia Utility-Scale PV Systems
DOI:
https://doi.org/10.69650/rast.2026.263014Keywords:
Sliding Window, Deep Learning, Utility-Scale Photovoltaic, Large-Scale Solar, Solar Power ForecastingAbstract
Accurate forecasting of solar power in utility-scale photovoltaic (USPV) systems is critical for grid stability but remains challenging due to meteorological variability and the large spatial scale of these systems. However, the choice of sliding window size in time-series forecasting remains underexplored. This study introduces a deep learning-based forecasting framework that systematically evaluates
the impact of sliding window size on forecasting accuracy using multivariate time-series data. The data collected from a 25 MWac USPV system in Malaysia between August 2022 and April 2023, comprises 5-minute interval measurements of solar irradiance, module temperature and solar power output. Multiple deep learning (DL) models, namely LSTM, CNN and GRU across window sizes ranging from 12 to 288 steps and forecasting horizons of 1 to 12 hours were investigated. Results show that a 144-step window consistently improves accuracy over conventional one-step input methods, with LSTM outperforming other models by achieving up to 23.1% RMSE reduction, 30.7% MAE reduction and a 8.6% increase in R² at 60 minutes forecasting horizon. This work emphasizes the importance of window size selection in optimizing forecasting accuracy for USPV systems and supporting renewable energy grid integration. By improving forecasting capabilities, this research is expected to provide critical insights to enhance renewable energy integration into the grid system.
References
Wilson, G. M., Al-Jassim, M., Metzger, W. K., Glunz, S. W., Verlinden, P., Xiong, G., Mansfield, L. M., Stanbery, B. J., Zhu, K., Yan, Y., Berry, J. J., Ptak, A. J., Dimroth, F., Kayes, B. M., Tamboli, A. C., Peibst, R., Catchpole, K., Reese, M. O., Klinga, C. S., Denholm, P., Morjaria, M., Deceglie, M. G., Freeman, J. M., Mikofski, M. A., Jordan, D. C., TamizhMani, G. and Sulas-Kern, D. B., The 2020 photovoltaic technologies roadmap. Journal of Physics D: Applied Physics. 53 (2020) 493001, doi: https://doi.org/10.1088/1361-6463/ab9c6a.
Zainuddin, H., Hanifah, M. M. M., Hussin, M. Z. and Muhammad, N. (2024). Cell temperature determination based on IEC61215: Solar photovoltaic experimental study of tropical Malaysia. Journal of Renewable Energy and Smart Grid Technology. 19 (2) 1–6, doi: https://doi.org/10.69650/rast.2024.258005.
Energy Commission. LSS Progress By Region (Q1 2024), <https://www.st.gov.my/eng//web/general/details/999> (2024).
Hammad, M., Khalil, S. and Mahmoud, I. M. Application of nonparametric ML on forecasting the production of a large-scale solar power plant: Kom-Ombo case study. Sustainable Computing: Informatics and Systems. 45 (2025) 101074.
Thansouphanh, B., Premrudeepreechacharn, S., Srirattanachaikul, W. and Ngamsanroaj, K., Predicting the temperature trend of a generator using RNN, GRU, and LSTM algorithms at Nam Ngum 1 hydropower plant in Laos. Journal of Renewable Energy and Smart Grid Technology. 20 (2025) 54–60, doi: https://doi.org/10.69650/rast.2025.260236.
Abumohsen, M., Owda, A. Y., Owda, M. and Abumihsan, A., Hybrid machine learning model combining of CNN-LSTM-RF for time series forecasting of solar power generation. e-Prime - Advances in Electrical Engineering. Electronics and Energy. 9 (2024) 100636, doi: https://doi.org/10.1016/j.prime.2024.100636.
Liu, C. H., Gu, J. C. and Yang, M. T., A simplified LSTM neural networks for one day-ahead solar power forecasting. IEEE Access. 9 (2021) 17174–17195, doi: https://doi.org/10.1109/ACCESS.2021.3053638.
Ozbek, A., Yildirim, A. and Bilgili, M., Deep learning approach for one-hour ahead forecasting of energy production in a solar-PV plant. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects. 44 (2022) 10173–10186, doi: https://doi.org/10.1080/15567036.2021.1924316.
Kumari, P. and Toshniwal, D., Long short term memory–convolutional neural network based deep hybrid approach for solar irradiance forecasting. Applied Energy. 295 (2021) 117061, doi: https://doi.org/10.1016/j.apenergy.2021.117061.
Bhatt, A., Ongsakul, W., Nimal Madhu, M. and Singh, J. G., Sliding window approach with first-order differencing for very short-term solar irradiance forecasting using deep learning models. Sustainable Energy Technologies and Assessments. 50 (2022) 101864, doi: https://doi.org/10.1016/j.seta.2021.101864.
Cican, G., Buturache, A.-N. and Silivestru, V., Predicting photovoltaic energy production using neural networks: Renewable integration in Romania. Processes. 13 (2025) 2219, doi: https://doi.org/10.3390/pr13072219.
du Plessis, A. A., Strauss, J. M. and Rix, A. J., Short-term solar power forecasting: Investigating the ability of deep learning models to capture low-level utility-scale photovoltaic system behaviour. Applied Energy. 285 (2021) 116395, doi: https://doi.org/10.1016/j.apenergy.2020.116395.
Xu, Y., Zheng, S., Zhu, Q., Wong, K. C., Wang, X. and Lin, Q., A complementary fused method using GRU and XGBoost models for long-term solar energy hourly forecasting. Expert Systems with Applications. 254 (2024) 124286, doi: https://doi.org/10.1016/j.eswa.2024.124286.
Kim, J., Kim, H., Kim, H. G., Lee, D. and Yoon, S., A comprehensive survey of deep learning for time series forecasting: Architectural diversity and open challenges. Artificial Intelligence Review. 58 (2025) 216, doi: https://doi.org/10.1007/s10462-025-11223-9.
Huang, X., Shi, J., Gao, B., Tai, Y., Chen, Z. and Zhang, J., Time series forecasting for hourly photovoltaic power using conditional generative adversarial network and Bi-LSTM. Energy. 246 (2022) 123403, doi: https://doi.org/10.1016/j.energy.2022.123403.
Boubaker, S., Benghanem, M., Mellit, A., Lefza, A., Kahouli, O. and Kolsi, L., Deep neural networks for predicting solar radiation at Hail Region, Saudi Arabia. IEEE Access. 9 (2021) 36719–36729, doi: https://doi.org/10.1109/ACCESS.2021.3062205.
Yang, D., Li, M., Guo, J. and Du, P., An attention-based multi-input LSTM with sliding window-based two-stage decomposition for wind speed forecasting. Applied Energy. 375 (2024) 124057, doi: https://doi.org/10.1016/j.apenergy.2024.124057.
Cui, X., Yu, X., Niu, H., Niu, D. and Liu, D., A novel data-driven multi-step wind power point-interval prediction framework integrating sliding window-based two-layer adaptive decomposition and multi-objective optimization for balancing prediction accuracy and stability. Applied Energy. 397 (2025) 126348, doi: https://doi.org/10.1016/j.apenergy.2025.126348.
Rana, M. and Rahman, A., Multiple steps ahead solar photovoltaic power forecasting based on univariate machine learning models and data re-sampling. Sustainable Energy, Grids and Networks. 21 (2020) 100286, doi: https://doi.org/10.1016/j.segan.2019.100286.
Sharadga, H., Hajimirza, S. and Balog, R. S., Time series forecasting of solar power generation for large-scale photovoltaic plants. Renewable Energy. 150 (2020) 797–807, doi: https://doi.org/10.1016/j.renene.2019.12.131.
Schwertman, N. C., Owens, M. A. and Adnan, R., A simple more general boxplot method for identifying outliers. Computational Statistics & Data Analysis. 47 (2004) 165–174, doi: https://doi.org/10.1016/j.csda.2003.10.012.
Xiao, Z., Huang, X., Liu, J., Li, C. and Tai, Y., A novel method based on time series ensemble model for hourly photovoltaic power prediction. Energy. 276 (2023) 127542, doi: https://doi.org/10.1016/j.energy.2023.127542.
Wang, J., Guo, L., Zhang, C., Song, L., Duan, J. and Duan, L., Thermal power forecasting of solar power tower system by combining mechanism modeling and deep learning method. Energy. 208 (2020) 118403, doi: https://doi.org/10.1016/j.energy.2020.118403.
Ahmed, R., Sreeram, V., Togneri, R., Datta, A. and Arif, M. D., Computationally expedient photovoltaic power forecasting: A LSTM ensemble method augmented with adaptive weighting and data segmentation technique. Energy Conversion and Management. 258 (2022) 115563, doi: https://doi.org/10.1016/j.enconman.2022.115563.
Abumohsen, M., Owda, A. Y. and Owda, M., Electrical load forecasting using LSTM, GRU, and RNN algorithms. Energies. 16 (2023) 2283, doi: https://doi.org/10.3390/en16052283.
Jia, P., Zhang, H., Liu, X. and Gong, X., Short-term photovoltaic power forecasting based on VMD and ISSA-GRU. IEEE Access. 9 (2021) 105939–105950, doi: https://doi.org/10.1109/ACCESS.2021.3099169.
Kabir, S., Islam, R. U., Hossain, M. S. and Andersson, K., An integrated approach of belief rule base and convolutional neural network to monitor air quality in Shanghai. Expert Systems with Applications. 206 (2022) 117905, doi: https://doi.org/10.1016/j.eswa.2022.117905.
Sulistya Kurniawan, A. D., Imanda Purnama, Y., Wicaksono, A. B. and Mahmudah, H. Comparative analysis and optimization of deep learning models for object detection using grid search hyperparameter tuning. in 2024 International Electronics Symposium (IES). (2024), 587–592, doi: https://doi.org/10.1109/IES63037.2024.10665797.
Quan, S. J., Comparing hyperparameter tuning methods in machine learning based urban building energy modeling: A study in Chicago. Energy and Buildings. 317 (2024) 114353, doi: https://doi.org/10.1016/j.enbuild.2024.114353.
Passos, D. and Mishra, P., A tutorial on automatic hyperparameter tuning of deep spectral modelling for regression and classification tasks. Chemometrics and Intelligent Laboratory Systems. 223 (2022) 104520, doi: https://doi.org/10.1016/j.chemolab.2022.104520.
Chicco, D., Warrens, M. J. and Jurman, G., The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation. PeerJ Computer Science. 7 (2021) e623, doi: https://doi.org/10.7717/PEERJ-CS.623.
Barhmi, K., Heynen, C., Golroodbari, S. and van Sark, W., A review of solar forecasting techniques and the role of artificial intelligence. Solar. 4 (2024) 101–139, doi: https://doi.org/10.3390/solar4010005.
Mei, F., Gu, J., Pan, S., Zhong, L., Lu, J., Jiang, Y. and Zheng, J., Day-ahead nonparametric probabilistic forecasting of photovoltaic power generation based on the LSTM-QRA ensemble model. IEEE Access. 8 (2020) 166138–166149, doi: https://doi.org/10.1109/ACCESS.2020.3021581.
Nguyen, N. Q., Bui, L. D., Doan, B. V., Sanseverino, E. R., Di Cara, D. and Nguyen, Q. D., A new method for forecasting energy output of a large-scale solar power plant based on long short-term memory networks: A case study in Vietnam. Electric Power Systems Research. 199 (2021) 107427, doi: https://doi.org/10.1016/j.epsr.2021.107427.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 School of Renewable Energy and Smart Grid Technology (SGtech)
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
All copyrights of the above manuscript, including rights to publish in any media, are transferred to the SGtech.
The authors retain the following rights;
1. All proprietary rights other than copyright.
2. Re-use of all or part of the above manuscript in their work.
3. Reproduction of the above manuscript for author’s personal use or for company/institution use provided that
(a) prior permission of SGtech is obtained,
(b) the source and SGtech copyright notice are indicated, and
(c) the copies are not offered for sale.
