A Hybrid Forecasting Approach for Solar Power Generation in Smart Grids Using Long Short-Term Memory and Autoregressive Integrated Moving Average

Year 2025, Volume: 5 Issue: 3, 216 - 224, 30.10.2025

Radhakrishnan Kanthavel Dhaya Ramakrishnan

https://doi.org/10.5152/tepes.2025.25021

https://izlik.org/JA74PL92WP

Abstract

An accurate solar energy forecast is important for the efficient operation of smart grids, especially with the increasing penetration of renewable energy sources. This paper proposes a hybrid forecasting approach that combines long short-term memory (LSTM) neural networks with autoregressive integrated moving average (ARIMA) models to improve the accuracy of short-term solar energy predictions. While ARIMA effectively captures linear temporary dependence, LSTM networks are powerful in nonlinear and long-distance pattern modeling. By integrating these two models, the proposed hybrid approach takes advantage of their complementary strengths to stop and address nonlinearity. The model is trained and tested on real-world solar power data collected from the gridconnected photovoltaic system. The evaluation metrics, such as mean absolute error, root mean squared error, and mean absolute percentage error, perform better than stand-alone ARIMA and LSTM models in the hybrid model, outpacing accuracy. Results outline the ability of hybrid intelligent models to increase the prediction of solar energy, contributing to more stable and reliable smart grid operations.

Keywords

Autoregressive integrated moving average (ARIMA) , hybrid models , long short-term memory , solar power forecasting , smart grids

References

• 1. G. E. P. Box, G. M. Jenkins, and G. C. Reinsel, Time Series Analysis: Forecasting and Control, 5th ed. Hoboken, NJ, USA: John Wiley & Sons, 2015.
• 2. S. Hochreiter, and J. Schmidhuber, “Long short-term memory,” Neural Comput., vol. 9, no. 8, pp. 1735–1780, Nov. 1997.
• 3. J. Wang, and D. Srinivasan, “A review of artificial intelligence-based building energy use prediction: Contrasting the capabilities of single and hybrid models,” Renew. Sustain. Energy Rev., vol. 75, pp. 796–808, Aug. 2017.
• 4. G. Zhang, B. E. Patuwo, and M. Y. Hu, “Forecasting with artificial neural networks: The state of the art,” Int. J. Forecast, vol. 14, no. 1, pp. 35–62, Mar. 1998.
• 5. A. Zameer, M. S. Khan, A. A. Naveed, and M. W. Ashraf, “Short-term solar power forecasting using hybrid ARIMA and LSTM models,” Energy Rep., vol. 6, pp. 446–454, Nov. 2020.
• 6. A. Mellit, and S. A. Kalogirou, “Artificial intelligence techniques for photovoltaic applications: A review,” Prog. Energy Combust. Sci., vol. 34, no. 5, pp. 574–632, Oct. 2008.
• 7. A. Khosravi, S. Nahavandi, D. Creighton, and A. F. Fathi, “Comprehensive review of neural network-based prediction intervals and new advances,” IEEE Trans. Neural Netw., vol. 22, no. 9, pp. 1341–1356, Sep. 2011.
• 8. J. Zeng, and W. Qiao, “Short-term solar power prediction using a support vector machine,” Renew. Energy, vol. 52, pp. 118–127, Apr. 2013.
• 9. R. Ahmed, V. Sreeram, Y. Mishra, and M. D. Arif, “A review and evaluation of the state-of-the-art in PV solar power forecasting: Techniques and optimization,” Renew. Sustain. Energy Rev., vol. 124, p. 109792, May 2020.
• 10. Y. Zhang, H. Zhang, L. Xie, and J. Wang, “Short-term photovoltaic power forecasting using a weighted neural network,” Energy Procedia, vol. 12, pp. 894–900, 2011.
• 11. A. T. Tascikaraoglu, and M. Uzunoglu, “A review of combined approaches for prediction of energy consumption in residential and commercial buildings,” Renew. Sustain. Energy Rev., vol. 34, pp. 36–50, Jun. 2014.
• 12. C. Zhu, Q. Liu, and S. Fan, “Hybrid forecasting model of photovoltaic power output using fuzzy logic and neural networks,” Sol. Energy, vol. 159, pp. 55–70, Jan. 2018.
• 13. A. Das, B. Roy, and A. Sengupta, “Short-term solar power prediction using hybrid convolutional LSTM networks,” Appl. Energy, vol. 304, p. 117843, Feb. 2022.
• 14. M. Z. Khan, M. H. Yousaf, M. Hussain, and I. Hussain, “Solar power prediction using deep LSTM networks and data pre-processing techniques,” Energy Rep., vol. 8, pp. 1421–1430, 2022.
• 15. Y. Zhang, S. Wang, D. Liu, and Y. Sun, “Hybrid Arima-LSTM model for wind speed forecasting with improved performance,” Renew. Energy, vol. 203, pp. 503–514, 2023.
• 16. T. Ahmad, H. Chen, and Y. Guo, “Deep learning-based hybrid model for photovoltaic power forecasting using CNN and BiLSTM,” Appl. Energy, vol. 309, p. 118358, 2022.
• 17. Z. Shao, J. Zhuang, Y. Lin, and X. Wang, “A novel hybrid method for dayahead solar forecasting using wavelet transform and GRU network,” Energy, vol. 263, p. 125623, 2023.
• 18. R. Mahapatra, R. K. Behera, and J. Nayak, “Energy forecasting using hybrid deep neural models: A review,” Mater. Today Proc., vol. 62, pp. 3212–3217, 2022.
• 19. X. Fang, Z. Li, X. Li, and Q. Li, “Day-ahead solar power forecasting using attention-based LSTM and meteorological features,” IEEE Access, vol. 10, pp. 5093–5102, 2022.
• 20. A. Sahoo, D. R. Patel, and S. Pal, “Short-term solar PV forecasting using CNN-LSTM hybrid deep learning for real-time microgrid control,” Appl. Energy, vol. 310, p. 118592, 2022.
• 21. Y. Wang, L. Zhang, and K. Chen, “Solar power forecasting using a Transformer model with multivariate exogenous variables,” IEEE Trans. Sus- tain. Energy, vol. 15, no. 1, pp. 234–244, 2023.
• 22. H. Chen, R. Zhang, and Q. Lin, “Adaptive attention-based GRU model for photovoltaic power forecasting under uncertain weather,” Energy Rep., vol. 8, pp. 1422–1434, 2022.
• 23. S. Ali, F. Mehmood, and N. Yousaf, “Comparative study of residual hybrid models and black-box neural approaches in solar forecasting,” Renew. Energy, vol. 201, pp. 566–578, 2022.
• 24. J. Kaur, P. Singh, and R. Goyal, “Deployment readiness of hybrid solar power forecasting models in utility-scale smart grids,” Sustain. Energy Grids Netw., vol. 27, p. 100536, 2021.
• 25. M. Hossain, N. Roy, and T. Yamashita, “High-resolution solar power prediction using attention-based LSTM with transfer learning,” Energy AI, vol. 9, p. 100186,