Hybrid SARIMA-NARNET modeling for accurate solar irradiance forecasting and energy efficiency optimization in tropical regions: A case study of Java-Bali

Sugeng Purwanto; Budi Sudiarto; Rinaldy Dalimi; Sugeng Purwanto; Budi Sudiarto; Rinaldy Dalimi

doi:10.3934/energy.2026023

AIMS Energy

2026, Volume 14, Issue 3: 548-570. doi: 10.3934/energy.2026023

Previous Article Next Article

Research article

Hybrid SARIMA-NARNET modeling for accurate solar irradiance forecasting and energy efficiency optimization in tropical regions: A case study of Java-Bali

Department of Electrical Engineering, Universitas Indonesia, Depok 16424, Indonesia

Received: 06 December 2025 Revised: 27 April 2026 Accepted: 08 May 2026 Published: 21 May 2026

Accurate solar irradiance forecasting is essential for renewable energy in tropical regions like Java-Bali, where weather variability poses major challenges. This study compared statistical (SARIMA) and neural network-based models (LSTM, GRU, NARNET, WNN), along with hybrid approaches, to identify the most effective prediction method. SARIMA was selected for its ability to capture consistent seasonal and linear trends, while NNVs model nonlinear relationships, especially in unstable weather. The proposed models were benchmarked against persistence and ARIMA baselines. The hybrid SARIMA-NARNET model achieved superior accuracy, with an MAE of 1.9287 W/m², RMSE of 2.5197 W/m², and a remarkably low MAPE of 0.3084%. Additionally, the DCL strategy demonstrated adaptive energy management, yielding daily energy savings of 16%–17% compared to static methods, with even greater efficiency at extreme confidence levels. These findings highlight the potential of hybrid modeling and adaptive control for optimizing solar energy use in tropical climates.
- solar irradiation forecasting,
- SARIMA-NNV hybrid model,
- solar power plant,
- Java-Bali,
- renewable energy
Citation: Sugeng Purwanto, Budi Sudiarto, Rinaldy Dalimi. Hybrid SARIMA-NARNET modeling for accurate solar irradiance forecasting and energy efficiency optimization in tropical regions: A case study of Java-Bali[J]. AIMS Energy, 2026, 14(3): 548-570. doi: 10.3934/energy.2026023

Related Papers:

Abstract

Accurate solar irradiance forecasting is essential for renewable energy in tropical regions like Java-Bali, where weather variability poses major challenges. This study compared statistical (SARIMA) and neural network-based models (LSTM, GRU, NARNET, WNN), along with hybrid approaches, to identify the most effective prediction method. SARIMA was selected for its ability to capture consistent seasonal and linear trends, while NNVs model nonlinear relationships, especially in unstable weather. The proposed models were benchmarked against persistence and ARIMA baselines. The hybrid SARIMA-NARNET model achieved superior accuracy, with an MAE of 1.9287 W/m², RMSE of 2.5197 W/m², and a remarkably low MAPE of 0.3084%. Additionally, the DCL strategy demonstrated adaptive energy management, yielding daily energy savings of 16%–17% compared to static methods, with even greater efficiency at extreme confidence levels. These findings highlight the potential of hybrid modeling and adaptive control for optimizing solar energy use in tropical climates.

References

[1]	Mubarak S, Impron, June DT (2019) Efficiency of solar radiation utilization and soybean (Glycine max L.) response to reflective mulch application. J Agron Indones 46: 247–253. https://doi.org/10.24831/jai.v46i3.18220 doi: 10.24831/jai.v46i3.18220
[2]	Mukhlisoti B, Garniwa I, (2023) Analysis of improvements to the automatic generation control (AGC) frequency regulation system in the Java Madura Bali system for intermittent new and renewable energy (EBT) interconnection. JETBIS 2: 470–489. https://doi.org/10.57185/jetbis.v2i7.49 doi: 10.57185/jetbis.v2i7.49
[3]	Halidah H, Prastawa A, Ariffikriyadi Z, et al. (2022) The geographic smoothing potential of solar irradiation in Java, Madura, and Bali electric power system. In 2022 5th International Conference on Power Engineering and Renewable Energy (ICPERE), Bandung, Indonesia, 1–5. https://doi.org/10.1109/ICPERE56870.2022.10037361
[4]	Anuradha K, Erlapally D, Karuna G, et al. (2021) Analysis of solar power generation forecasting using machine learning techniques. In 3rd International Conference on Design and Manufacturing Aspects for Sustainable Energy (ICMED-ICMPC 2021), 01163. https://doi.org/10.1051/e3sconf/202130901163
[5]	Poddar S, Kay M, Prasad A, et al. (2023) Changes in solar resource intermittency and reliability under Australia's future warmer climate. Sol Energy 266: 112039. https://doi.org/10.1016/j.solener.2023.112039 doi: 10.1016/j.solener.2023.112039
[6]	Sigismondi C, Morcos AB (2022) Long term variations of solar radius. Gen Relativ Gravit 43: 1197–1202. https://doi.org/10.1007/s10714-010-0958-8 doi: 10.1007/s10714-010-0958-8
[7]	Almarzooqi AM, Maalouf M, El-Fouly THM, et al. (2024) A hybrid machine-learning model for solar irradiance forecasting. Clean Energy 8: 100–110. https://doi.org/10.1093/ce/zkad075 doi: 10.1093/ce/zkad075
[8]	Jamil B, Irshad K, Algahtani A, et al. (2019) On the calibration and applicability of global solar radiation models based on temperature extremities in India. Environ Prog Sustainable Energy 39: e13236. https://doi.org/10.1002/ep.13236 doi: 10.1002/ep.13236
[9]	Saud S, Jamil B, Upadhyay Y, et al. (2020) Performance improvement of empirical models for estimation of global solar radiation in India: A k-fold cross-validation approach. Sustainable Energy Technol Assess 40: 100768. https://doi.org/10.1016/j.seta.2020.100768 doi: 10.1016/j.seta.2020.100768
[10]	Sivarajan S, Sundarsingh Jebaseelan SD (2025) A hybrid CNN-RNN based energy consumption forecasting for smart grids in Industries. Int J Renewable Energy Res 15: 172–180. https://doi.org/10.20508/ijrer.v15i1.14176.g9031 doi: 10.20508/ijrer.v15i1.14176.g9031
[11]	Kushwaha V, Pindoriya NM (2017) Very short-term solar PV generation forecast using SARIMA model: A case study. In 2017 7th International Conference on Power Systems (ICPS), Pune, India, 430–435. https://doi.org/10.1109/ICPES.2017.8387332
[12]	Kumari S, Muthulakshmi P (2023) SARIMA model: An efficient machine learning technique for weather forecasting. Procedia Comput Sci 235: 656–670. https://doi.org/10.1016/j.procs.2024.04.064 doi: 10.1016/j.procs.2024.04.064
[13]	Barus DH, Dalimi R (2021) Determining optimal operating reserves toward wind power penetration in Indonesia based on hybrid artificial intelligence. IEEE Access 9: 165173–165183. https://doi.org/10.1109/ACCESS.2021.3135261 doi: 10.1109/ACCESS.2021.3135261
[14]	Alharbi FR, Csala D (2022) A seasonal autoregressive integrated moving average with exogenous factors (SARIMAX) forecasting model-based time series approach. Inventions 7: 94. https://doi.org/10.3390/inventions7040094 doi: 10.3390/inventions7040094
[15]	Neshat M, Nezhad MM, Mirjalili S, et al. (2023) Short-term solar radiation forecasting using hybrid deep residual learning and gated LSTM recurrent network with differential covariance matrix adaptation evolution strategy. Energy 278: 127701. https://doi.org/10.1016/j.energy.2023.127701 doi: 10.1016/j.energy.2023.127701
[16]	Gufron A, Garniwa PMP, Putera DA, et al. (2025) A preliminary LSTM-IDW model for spatiotemporal hourly solar radiation estimation in tropical regions. Renewable Sustainable Energy Trans 7: 100105. https://doi.org/10.1016/j.rset.2025.100105 doi: 10.1016/j.rset.2025.100105
[17]	Tatachar AV (2021) Comparative assessment of regression models based on model evaluation metrics. Int Res J Eng Technol 8: 853–860. Available from: https://www.irjet.net/archives/V8/i9/IRJET-V8I9127.pdf.
[18]	Ge Y, Nan Y, Bai L (2019) A hybrid prediction model for solar radiation based on long short-term memory, empirical mode decomposition, and solar profiles for energy harvesting wireless sensor networks. Energies 12: 4762. https://doi.org/10.3390/en12244762 doi: 10.3390/en12244762
[19]	Wang Y, Yao Y, Zou Q, et al. (2024) Forecasting a short-term photovoltaic power model based on improved snake optimization, convolutional neural network, and bidirectional long short-term memory network. Sensors 24: 3897. https://doi.org/10.3390/s24123897 doi: 10.3390/s24123897
[20]	Sarmas E, Spiliotis E, Stamatopoulos E, et al. (2023) Short-term photovoltaic power forecasting using meta-learning and numerical weather prediction independent long short-term memory models. Renewable Energy 216: 118997. https://doi.org/10.1016/j.renene.2023.118997 doi: 10.1016/j.renene.2023.118997
[21]	Fang X, Han S, Li J. et al. (2023) A FCM-XGBoost-GRU model for short-term photovoltaic power forecasting based on weather classification. In 2023 5th Asia Energy and Electrical Engineering Symposium (AEEES), Chengdu, China, 1444–1449. https://doi.org/10.1109/AEEES56888.2023.10114292
[22]	Sianturi YCU (2018) Estimating the solar energy potential over Indonesia region using daily sunshine duration. Int J Sci Res 9: 1069–1073. https://doi.org/10.21275/SR20510190145 doi: 10.21275/SR20510190145
[23]	Sarjiya, Putranto LM, Irnawan R, et al. (2022) Assessing potential scenarios for achieving new and renewable energy targets in Java-Bali power system, Indonesia. Int J Energy Econ Policy 12: 502–515. https://doi.org/10.32479/ijeep.12852 doi: 10.32479/ijeep.12852
[24]	Ali H, Sudiarto B, Jufri FH, et al. (2024) Study of reducing transmission losses in Java-Bali system with the addition of capacitors along with optimal capacitor placement methods using quasi-dynamic simulation. Int J Electr Comput Biomed Eng 2: 1–18. https://doi.org/10.62146/ijecbe.v2i1.39 doi: 10.62146/ijecbe.v2i1.39
[25]	Rencana Usaha Penyediaan Tenaga Listrik (RUPTL) PT PLN (Persero) 2021–2030. Rencana Usaha Penyediaan Tenaga Listrik 2021–2030, 2019–2028. Available from: https://web.pln.co.id/stakeholder/ruptl.
[26]	Institute for Essential Services Reform (IESR) (2020) Indonesia Energy Transition Outlook 2025. Essential Concepts of Global Environmental Governance, 86–88. Available from: https://iesr.or.id/en/pustaka/indonesia-energy-transition-outlook-ieto-2025/.
[27]	Wisudawati N, Fijra R (2021) Analisis Efektivitas Penggunaan Energi Baru Dan Terbarukan Di Provinsi Sumatera Selatan Guna Mendukung REUN 2025. Integrasi : Jurnal Ilmiah Teknik Industri 6: 1–9. https://doi.org/10.32502/js.v6i1.3789
[28]	Saifurrohman MH, Hasyid MH, Putranto LM, et al. (2023) Battery energy storage systems reinforcement control strategy to enhanced the maximum integration of PV to generation systems. Results Eng 18: 101184. https://doi.org/10.1016/j.rineng.2023.101184 doi: 10.1016/j.rineng.2023.101184
[29]	PT PLN Energi Primer Indonesia (2023) Statistics Report 2023. Available from: https://cmsadmin.plnepi.co.id/storage/media/EPI-STATISTIK-2023-1709%20-%20Final%20Cetak.pdf.
[30]	Mohi Ud Din A, Gupta V (2023) Forecasting and prediction of solar energy in solar photovoltaic plants. Tuijin Jishu J Propuls Technol 44: 1457–1469. https://doi.org/10.52783/tjjpt.v44.i4.1080 doi: 10.52783/tjjpt.v44.i4.1080
[31]	Barhmi K, Heynen C, Golroodbari S, et al. (2024) A review of solar forecasting techniques and the role of artificial intelligence. Solar 4: 99–135. https://doi.org/10.3390/solar4010005 doi: 10.3390/solar4010005
[32]	Zheng K, Sun Z, Song Y, et al., (2025) Stochastic scenario generation methods for uncertainty in wind and photovoltaic power outputs: A comprehensive review. Energies 18: 503. https://doi.org/10.3390/en18030503 doi: 10.3390/en18030503
[33]	Fang T, Lahdelma R (2916) Evaluation of a multiple linear regression model and SARIMA model in forecasting heat demand for district heating system. Appl Energy 179: 544–552. https://doi.org/10.1016/j.apenergy.2016.06.133 doi: 10.1016/j.apenergy.2016.06.133
[34]	Nur Aziza V, Moh'd FH, Maghfiroh FA, et al. (2023) Performance comparison of sarima intervention and prophet models for forecasting the number of airline passenger at Soekarno-Hatta International Airport. BAREKENG J Math Appl 17: 2107–2120. https://doi.org/10.30598/barekengvol17iss4pp2107-2120 doi: 10.30598/barekengvol17iss4pp2107-2120
[35]	Hassani H, Mashhad LM, Royer-Carenzi M, et al. (2025) White noise and its misapplications: Impacts on time series model adequacy and forecasting. Forecasting 7: 8. https://doi.org/10.3390/forecast7010008 doi: 10.3390/forecast7010008
[36]	Hassani H, Marvian L, Yarmohammadi M, et al. (2024) Unraveling time series dynamics: Evaluating partial autocorrelation function distribution and its implications. Math Comput Appl 29: 58. https://doi.org/10.3390/mca29040058 doi: 10.3390/mca29040058
[37]	Sghir A, Iroshan A (2024) Unleashing the power of LSTMs: Advancements and applications in sequential data processing, 1–15. https://dx.doi.org/10.2139/ssrn.5032045
[38]	Kong Y, Wang Z, Nie Y, et al. (2024) Unlocking the power of LSTM for long term time series forecasting. Preprint: https://doi.org/10.48550/arXiv.2408.10006
[39]	Bousnguar H, Nadji L, Battou A (2023) Gated recurrent units (GRU) for time series forecasting in higher education. Int J Eng Res Technol 12: 152–154. Available from: https://www.ijert.org/research/gated-recurrent-units-gru-for-time-series-forecasting-in-higher-education-IJERTV12IS030091.pdf.
[40]	Wang X, Xu J, Shi W, et al. (2019) OGRU: An optimized gated recurrent unit neural network. J Phys Conf Ser 1325: 012089. https://doi.org/10.1088/1742-6596/1325/1/012089 doi: 10.1088/1742-6596/1325/1/012089
[41]	Kong X, Chen Z, Liu W, et al. (2025) Deep learning for time series forecasting: A survey. Int J Mach Learn Cyber 16: 5079–5112. https://doi.org/10.1007/s13042-025-02560-w doi: 10.1007/s13042-025-02560-w
[42]	Wu YK, Chen CR, Abdul Rahman H (2014) A novel hybrid model for short-term forecasting in PV power generation. Int J Photoenergy 2014: 569249. https://doi.org/10.1155/2014/569249 doi: 10.1155/2014/569249

Reader Comments

Your name:*

Email:*
© 2026 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)