An AI-Based Hybrid Prophet–LSTM Model for Forecasting and Financial Optimisation in Sustainable Energy Grids

Tajul Rosli Razak; Ahmed El Alaoui; Hasila Jarimi; Ahmad Zia Ul-Saufie; Mohammad Hafiz Ismail; Mohd Faris Mohd Fuzi

doi:10.65582/aifsc.2026.004

Authors

Tajul Rosli Razak Faculty of Computer and Mathematical Sciences, Universiti Teknologi MARA (UiTM), Shah Alam, Malaysia; Solar Energy Research Institute, The National University of Malaysia, Bangi, Malaysia; Department of Architecture and Built Environment, University of Nottingham, Nottingham NG7 2RD, United Kingdom. https://orcid.org/0000-0002-6389-8108
Ahmed El Alaoui Emines – School of Industrial Management, Mohammed VI Polytechnic University (UM6P), Ben Guerir, Morocco https://orcid.org/0000-0003-3821-5245
Hasila Jarimi Solar Energy Research Institute, The National University of Malaysia, Bangi, Malaysia; Department of Architecture and Built Environment, University of Nottingham, Nottingham NG7 2RD, United Kingdom https://orcid.org/0000-0003-0921-3283
Ahmad Zia Ul-Saufie Faculty of Computer and Mathematical Sciences, Universiti Teknologi MARA (UiTM), Shah Alam, Malaysia https://orcid.org/0000-0002-4084-6647
Mohammad Hafiz Ismail Faculty of Computer and Mathematical Sciences, Universiti Teknologi MARA (UiTM), Arau, Perlis, Malaysia https://orcid.org/0000-0001-5798-4926
Mohd Faris Mohd Fuzi Faculty of Computer and Mathematical Sciences, Universiti Teknologi MARA (UiTM), Arau, Perlis, Malaysia. https://orcid.org/0000-0002-4395-2736

DOI:

https://doi.org/10.65582/aifsc.2026.004

Keywords:

Hybrid Prophet–LSTM, Energy finance forecasting, Robust forecasting, Model benchmarking, Explainable artificial intelligence , Sustainable energy grids

Abstract

Accurate and reliable forecasting is critical for sustainable energy grid planning, as fluctuating demand, market volatility, and policy uncertainty pose significant challenges to both operational and financial stability. This paper proposes a novel AI-based hybrid forecasting framework that integrates Facebook Prophet, for interpretable long-term trend and seasonality decomposition, with Long Short-Term Memory (LSTM) networks, for modelling nonlinear short-term residual dynamics. This design explicitly addresses the dual requirement of transparency and high predictive accuracy in complex, non-stationary energy systems. The framework is evaluated using 18 years of historical financial data from Tenaga Nasional Berhad (TNB), Malaysia’s largest electricity utility, as a representative large-scale grid operation case study. Performance is benchmarked against ARIMA, standalone Prophet, and standalone LSTM models using RMSE, MAE, MAPE, and SMAPE, with statistical significance assessed via the Diebold–Mariano test and robustness examined under varying forecast horizons and noise perturbations. Results show that the proposed hybrid Prophet–LSTM model achieves up to 15% lower RMSE and MAPE than the best-performing baseline while maintaining stable performance under adverse conditions. The findings demonstrate that the proposed framework provides a robust, interpretable, and modular decision-support tool for utility operators, energy planners, and policymakers, enabling improved financial optimisation, tariff planning, and operational resilience in sustainable energy grid systems.

References

Avinash, G., Ramasubramanian, V., Ray, M., Paul, R.K., Godara, S., Nayak, G.H., Kumar, R.R., Manjunatha, B., Dahiya, S. and Iquebal, M.A. (2024). Hidden Markov guided deep learning models for forecasting highly volatile agricultural commodity prices. Applied Soft Computing, 158, p. 111557. DOI:10.1016/j.asoc.2024.111557.

Box, G.E., Jenkins, G.M., Reinsel, G.C. and Ljung, G.M. (2015). Time series analysis: forecasting and control. Hoboken: John Wiley & Sons.

Cheng, J., Deng, H., Sun, G., Guo, P. and Zhang, J. (2020). Application of ARIMA model in financial time series in stocks. Lecture Notes in Computer Science, 12239, pp. 232–243.

Cohen, J., Kaur, S., Li, Y., Kolter, J.Z. and Talwalkar, A. (2021) Gradient descent on neural networks typically occurs at the edge of stability. Proceedings of the 9th International Conference on Learning Representations (ICLR 2021).

Hamilton, J.D. (2010). Regime switching models. In: Macroeconometrics and Time Series Analysis, pp. 202–209.

He, C. (2023). A hybrid model based on multi-LSTM and ARIMA for time series forecasting. Proceedings of the 8th International Conference on Intelligent Computing and Signal Processing (ICSP 2023), pp. 612–616.

Hochreiter, S. and Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), pp. 1735–1780.

Hsu, M.W., Lessmann, S., Sung, M.C., Ma, T. and Johnson, J.E. (2016). Bridging the divide in financial market forecasting: Machine learners vs. financial economists. Expert Systems with Applications, 61, pp. 215–234.

Lago, J., De Ridder, F. and De Schutter, B. (2018). Forecasting spot electricity prices: Deep learning approaches and empirical comparison of traditional algorithms, Applied Energy, 221, pp. 386–405. DOI:10.1016/j.apenergy.2018.02.069.

Lim, B. and Zohren, S. (2021). Time-series forecasting with deep learning: A survey, Philosophical Transactions of the Royal Society A, 379(2194). DOI:10.1098/rsta.2020.0209.

Liu, H., Tian, H.Q. and Li, Y.F. (2012). Comparison of two new ARIMA-ANN and ARIMA-Kalman hybrid methods for wind speed prediction, Applied Energy, 98, pp. 415–424.

Liu, Q., Wang, Q. and Wang, X. (2025). RVFL-LSTM: A lightweight model with long-short term memory for time series, Knowledge-Based Systems, 309, 112896.

Makridakis, S., Spiliotis, E. and Assimakopoulos, V. (2018). Statistical and machine learning forecasting methods: Concerns and ways forward, PLOS ONE, 13(3), e0194889. DOI:10.1371/journal.pone.0194889.

Mohamed, E., Sirlantzis, K. and Howells, G. (2022). A review of visualisation-as-explanation techniques for convolutional neural networks and their evaluation, Displays, 73, 102239.

Mohammadi, M., Jamshidi, S., Rezvanian, A., Gheisari, M. and Kumar, A. (2024). Advanced fusion of MTM-LSTM and MLP models for time series forecasting: An application for forecasting the solar radiation, Measurement: Sensors, 33, 101179.

Nystrup, P., Madsen, H. and Lindström, E. (2017). Long memory of financial time series and hidden Markov models with time-varying parameters, Journal of Forecasting, 36(8), pp. 989–1002.

Oukhouya, H., Kadiri, H., El Himdi, K. and Guerbaz, R. (2024). Forecasting international stock market trends: XGBoost, LSTM, LSTM-XGBoost, and backtesting XGBoost models, Statistics, Optimization & Information Computing, 12(1), pp. 200–209.

Pan, Y.A., Guo, J., Chen, Y., Cheng, Q., Li, W. and Liu, Y. (2024). A fundamental diagram based hybrid framework for traffic flow estimation and prediction by combining a Markovian model with deep learning, Expert Systems with Applications, 238, 122219.

Parviz, L. and Ghorbanpour, M. (2024). Assimilation of PSO and SVR into an improved ARIMA model for monthly precipitation forecasting, Scientific Reports, 14(1).

Pascanu, R., Mikolov, T. and Bengio, Y. (2013). On the difficulty of training recurrent neural networks, in Proceedings of the 30th International Conference on Machine Learning (ICML 2013), pp. 2347–2355.

Peng, J., Deng, C. and Chen, Y.C. (2023). Application of ARIMA time series model in stock data prediction, Smart Innovation, Systems and Technologies, 341, pp. 223–232.

Peng, Z., Khan, F.U., Khan, F., Shaikh, P.A., Yonghong, D., Ullah, I. and Ullah, F. (2021). An application of hybrid models for weekly stock market index prediction: Empirical evidence from SAARC countries, Complexity, 2021(1)

Rabiner, L.R. (1989). A tutorial on hidden Markov models and selected applications in speech recognition, Proceedings of the IEEE, 77(2), pp. 257–286.

Rizvi, M.F. (2024). ARIMA model time series forecasting, International Journal for Research in Applied Science and Engineering Technology, 12(5), pp. 3782–3785.

Shen, Z., Wan, Q. and Leatham, D.J. (2021). Bitcoin return volatility forecasting: A comparative study between GARCH and RNN, Journal of Risk and Financial Management, 14(7).

Taylor, S.J. and Letham, B. (2018). Forecasting at scale, The American Statistician, 72(1), pp. 37–45.

Thomson, M.E., Pollock, A.C., Önkal, D. and Gönül, M.S. (2019). Combining forecasts: Performance and coherence, International Journal of Forecasting, 35(2), pp. 474–484.

Torres, J.F., Hadjout, D., Sebaa, A., Martínez-Álvarez, F. and Troncoso, A. (2021). Deep learning for time series forecasting: A survey, Big Data, 9(1), pp. 3–21.

Wang, X., Tang, H., Liang, X., Zeng, Y., Qin, H. and Hu, H. (2025). RH-SAVis: Visual analytics of spatiotemporal situation awareness for ride-hailing multi-source data, Journal of Visualization, pp. 1–18.

Weron, R. (2014). Electricity price forecasting: A review of the state-of-the-art with a look into the future, International Journal of Forecasting, 30(4), pp. 1030–1081.

Xu, K., Chen, L., Patenaude, J.M. and Wang, S. (2024). RHINE: A regime-switching model with nonlinear representation for discovering and forecasting regimes in financial markets, in Proceedings, pp. 526–534.

Zhang, P.G. (2003). Time series forecasting using a hybrid ARIMA and neural network model, Neurocomputing, 50, pp. 159–175.

Zheng, H., Wu, J., Song, R., Guo, L. and Xu, Z. (2024). Predicting financial enterprise stocks and economic data trends using machine learning time series analysis, Applied and Computational Engineering, 87(1), pp. 26–32.