Accurate PV Power Forecasting Using a Lightweight LSTM Model: A Case Study

Document Type : Original Research

Authors

Mechanics of Biosystems Engineering Department, College of Aburaihan, University of Tehran, Tehran, Iran.

10.22103/bbr.2025.25811.1132

Abstract

This study presents a significant advancement in short-term photovoltaic (PV) power forecasting through the development and validation of a simple yet highly effective LSTM-based model tailored to address the operational demands of renewable energy integration. By harnessing a meticulously curated dataset and employing rigorous feature selection, the model achieved exceptional performance metrics an R² score of 0.9212 and an RMSE of 0.0650 on unseen data outperforming benchmark models such as CatBoost and GBR. These outcomes affirm the model's capacity to capture temporal dependencies in PV generation data while maintaining computational efficiency, making it well suited for real-time energy management applications. However, limitations such as dependence on high-quality input data and untested resilience under extreme weather conditions suggest areas for refinement. Future research could enhance the model by incorporating probabilistic forecasting, lightweight attention mechanisms, or transfer learning to improve adaptability across diverse geographic and climatic contexts. Ultimately, this work contributes a robust, practical tool to the evolving landscape of smart grid technologies, supporting the global transition toward sustainable energy systems with improved forecasting precision and scalability.

Keywords

References

Abdullah, W., Elmasry, A., & Tolba, A. (2024). Hybrid attention-enhanced deep learning for accurate hourly energy consumption forecasting. Information Sciences with Applications, 3, 74-83. https://doi.org/10.61356/j.iswa.2024.3314
Ahmed, I., Khan, M. A. U. H., Islam, M., Hasan, M. S., Jakir, T., Hossain, A., Abed, J., Hasanuzzaman, M., Shatyi, S. S., & Hasnain, K. N. (2025). Optimizing Solar Energy Production in the USA: Time-Series Analysis Using AI for Smart Energy Management. Journal of Posth umanism, 5(6), 3396-3423. https://doi.org/10.63332/joph.v5i6.2457
Alexakos, A., Amaxilatis, D., & Zaroliagis, C. (2022). Photovoltaic energy production forecasting and operational analytics: a real-world study. In 2022 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops) (pp. 439-444). IEEE.‏
An, Y., Dang, K., Shi, X., Jia, R., Zhang, K., & Huang, Q. (2021). A probabilistic ensemble prediction method for PV power in the nonstationary period. Energies, 14(4), 859. https://doi.org/10.3390/en14040859
Araya, F., Long-Ha, D., Eddine, M. D., & Al Shakarchi, F. (2023, October). Optimal Energy Management System Using Probabilistic Day-ahead Forecasting. In 2023 IEEE PES Innovative Smart Grid Technologies Europe (ISGT EUROPE) (pp. 1-5). IEEE.‏
Balal, A. T., Jafarabadi, Y. P. T., Demir, A. T., Igene, M. T., Giesselmann, M. T., & Bayne, S. T. (2023). Forecasting solar power generation utilizing machine learning models in Lubbock. Emerging Science Journal, 7(4), 1052-1062. https://doi.org/10.28991/ESJ-2023-07-04-02
Bracale, A., Carpinelli, G., & De Falco, P. (2016). A probabilistic competitive ensemble method for short-term photovoltaic power forecasting. IEEE Transactions on Sustainable Energy, 8(2), 551-560. https://doi.org/10.1109/TSTE.2016.2610523
Chicco, D., Warrens, M. J., & Jurman, G. (2021). The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation. Peerj computer science, 7, e623. https://doi.org/10.7717/peerj-cs.623
Di Leo, P., Ciocia, A., Malgaroli, G., & Spertino, F. (2025). Advancements and Challenges in Photovoltaic Power Forecasting: A Comprehensive Review. Energies, 18(8), 2108. https://doi.org/10.3390/en18082108
Fan, Y., Ma, Z., Tang, W., Liang, J., & Xu, P. (2024). Using crested Porcupine optimizer algorithm and CNN-LSTM-Attention model combined with deep learning methods to enhance short-term power forecasting in PV generation. Energies, 17(14), 3435. https://doi.org/10.3390/en17143435
Fraga-Hurtado, I., Gómez-Sarduy, J. R., García-Sánchez, Z., Hernández-Herrera, H., Silva-Ortega, J. I., & Reyes-Calvo, R. (2025). Advanced Multivariate Models Incorporating Non-Climatic Exogenous Variables for Very Short-Term Photovoltaic Power Forecasting. Electricity, 6(2), 29. https://doi.org/10.3390/electricity6020029
Hapsari, G. I., Munadi, R., Erfianto, B., & Irawati, I. D. (2025). Feature Selection Using Pearson Correlation for Ultra-Wideband Ranging Classification. Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi), 9(2), 209-217. https://doi.org/10.29207/resti.v9i2.6281
Izdebski, W., & Kosiorek, K. (2023). Analysis and evaluation of the possibility of electricity production from small photovoltaic installations in Poland. Energies, 16(2), 944. https://doi.org/10.3390/en16020944
Jailani, N. L. M., Dhanasegaran, J. K., Alkawsi, G., Alkahtani, A. A., Phing, C. C., Baashar, Y., Capretz, L. F., Al-Shetwi, A. Q., & Tiong, S. K. (2023). Investigating the power of LSTM-based models in solar energy forecasting. Processes, 11(5), 1382.
Jakoplić, A., Franković, D., Havelka, J., & Bulat, H. (2023). Short-term photovoltaic power plant output forecasting using sky images and deep learning. Energies, 16(14), 5428. https://doi.org/10.3390/en16145428
Liu, C., Li, M., Yu, Y., Wu, Z., Gong, H., & Cheng, F. (2022). A review of multitemporal and multispatial scales photovoltaic forecasting methods. IEEE access, 10, 35073-35093. https://doi.org/10.1109/ACCESS.2022.3162206
Mazen, F. M. A., Shaker, Y., & Abul Seoud, R. A. (2023). Forecasting of solar power using GRU–temporal fusion transformer model and DILATE loss function. Energies, 16(24), 8105. https://doi.org/10.3390/en16248105
Praveenraj, D. D. W., Madeswaran, A., Pastariya, R., Sharma, D., Abootharmahmoodshakir, K., & Dhablia, A. (2024). Machine learning integration for enhanced solar power generation forecasting. In E3S Web of Conferences (Vol. 540, p. 04007). EDP Sciences.‏ https://doi.org/10.1051/e3sconf/202454004007
Rezvani, Z., Rezvani, F., & Arslan, S. (2022). Designing, Simulating and Technical Analysis of a 2 MW On-grid Photovoltaic System for Agricultural Applications. Biomechanism and Bioenergy Research, 1(2), 1-6. https://doi.org/10.22103/bbr.2022.20681.1036
Robeson, S. M., & Willmott, C. J. (2023). Decomposition of the mean absolute error (MAE) into systematic and unsystematic components. PloS one, 18(2), e0279774. https://doi.org/10.1371/journal.pone.0279774
Sabri, M., & El Hassouni, M. (2023). Photovoltaic power forecasting with a long short-term memory autoencoder networks. Soft Computing, 27(15), 10533-10553. https://doi.org/10.1007/s00500-023-08497-y
Salman, D., Direkoglu, C., Kusaf, M., & Fahrioglu, M. (2024). Hybrid deep learning models for time series forecasting of solar power. Neural Computing and Applications, 36(16), 9095-9112. https://doi.org/10.1007/s00521-024-09558-5
Sharma, J., Soni, S., Paliwal, P., Saboor, S., Chaurasiya, P. K., Sharifpur, M., Khalilpoor, N., & Afzal, A. (2022). A novel long term solar photovoltaic power forecasting approach using LSTM with Nadam optimizer: A case study of India. Energy Science & Engineering, 10(8), 2909-2929.
Sleiman, A., & Su, W. (2024). Combined K-means clustering with neural networks methods for PV short-term generation load forecasting in electric utilities. Energies, 17(6), 1433. https://doi.org/10.3390/en17061433
Stoliarov, O. (2024). Big Data technologies in the process of forecasting electricity generation from solar photovoltaic power plants. Вісник Черкаського державного технологічного університету. Технічні науки, 29(2). 79-92.https://doi.org/10.62660/bcstu/2.2024.79
Suyambu, M. R., Vishwakarma, P. K., & Shrivastava, V. (2024, December). Improving Energy Efficiency Through Machine Learning-Based Load Forecasting in power systems. In 2024 International Conference on Communication, Control, and Intelligent Systems (CCIS) (pp. 1-6). IEEE.‏ https://doi.org/10.1109/CCIS63231.2024.10931928
Willmott, C. J., & Matsuura, K. (2005). Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Climate research, 30(1), 79-82. https://doi.org/10.3354/cr030079
Wu, Y., Gong, G., & Zhu, C. (2024, September). Electricity demand forecasting based on hybrid BiGRU-BiLSTM model prediction framework with dual time domain decomposition. In 2024 5th International Conference on Big Data & Artificial Intelligence & Software Engineering (ICBASE) (pp. 789-793). IEEE.‏ https://doi.org/10.1109/ICBASE63199.2024.10761991
Yadav, S., Rajput, P., Balasubramanian, P., Liu, C., Li, F., & Zhang, P. (2025). Machine learning-driven prediction of biochar adsorption capacity for effective removal of Congo red dye. Carbon Research, 4(1), 11. https://doi.org/10.1007/s44246-024-00168-3
Zhao, Z., Zhai, M., Li, G., Gao, X., Song, W., Wang, X., Ren, H., Cui, Y., Qiao, Y., & Ren, J. (2023). Study on the prediction effect of a combined model of SARIMA and LSTM based on SSA for influenza in Shanxi Province, China. BMC Infectious Diseases, 23(1), 71. https://doi.org/10.1186/s12879-023-08025-1
Biomechanism and Bioenergy Research
Volume 4, Issue 3 - Serial Number 9
September 2025
Pages 68-80

Files

History

  • Receive Date: 23 August 2025
  • Revise Date: 02 September 2025
  • Accept Date: 09 September 2025

Share

How to cite

Statistics