THERMAL SCIENCE

International Scientific Journal

Branka D GvozdenacORCID

,

Dušan D Gvozdenac

ENHANCING ELECTRICITY MIX FORECASTING WITH ARTIFICIAL NEURAL NETWORKS: A CASE STUDY OF THE EU27

ABSTRACT
Electric power systems are a cornerstone of the European Union's strategy to achieve climate neutrality by 2050. This study introduces a custom-built, transparent, and user-friendly model that leverages artificial neural networks to forecast the electricity mix based on historical generation data. The model accommodates a range of generation technologies, including coal, natural gas, other fossil fuels, nuclear, hydroelectric, wind, solar photovoltaics, bioenergy, and other renewables. In addition to the Business-as-Usual scenario, three alternative pathways are examined, each aligned with different strategic, technical, and policy objectives. The model estimates plant capacities required to meet projected electricity demand in 2050 while considering direct capital expenditures for new constructions and decommissioning. Results suggest that achieving the EU27's greenhouse gas reduction targets by 2050 will be extremely challenging under the current trajectory. A successful transition demands substantial adjustments in energy strategy at both EU and member state levels, underpinned by a comprehensive framework that integrates technical, economic, social, and environmental factors.
KEYWORDS
Electricity Mix, Near Carbon Neutrality by 2050, Techno-Economic Modeling, artificial neural networks, Energy Strategy
PAPER SUBMITTED: 2025-09-29
PAPER REVISED: 2026-01-09
PAPER ACCEPTED: 2026-01-23
PUBLISHED ONLINE: 2026-02-08
DOI REFERENCE: https://doi.org/10.2298/TSCI250929008G
REFERENCES
[1] "MARKAL," [Online]. Available: https://iea-etsap.org/index.php/etsap-tools/model-generators/markal
[2] "PLEXOS software," [Online]. Available: https://www.energyexemplar.com/plexos
[3] "EnergyPLAN," [Online]. Available: https://robinson-eb.eu/tools/res-potential-estimation/energyplan
[4] International Energy Agency (IEA) , [Online]. Available: https://www.iea.org/
[5] Energy Information Administration (EIA) , [Online]. Available: https://www.eia.gov/
[6] World Energy Council (WEC), [Online]. Available: https://www.worldenergy.org/
[7] "International Renewable Energy Agency (IRENA)," [Online]. Available: https://www.irena.org/
[8] "Center for Energy and Environmental Policy Research (CEEPR)," [Online]. Available: https://ceepr.mit.edu/
[9] "Carbon Capture and Storage Association (CCSA)," [Online]. Available: https://www.ccsassociation.org/
[10] STATISTA , [Online]. Available: https://www.statista.com/
[11] EMBER, [Online]. Available: https://ember-climate.org/data/
[12] A. Al-Dahidi, "Forecasting Solar Photovoltaic Power Production: A Comprehensive Review and Innovative Data-Driven Modeling Framework," Energies, vol. 17, 2024
[13] M. Alrashidi and S. Rahman, "Short-term photovoltaic power production forecasting based on novel hybrid data-driven models," Journal of Big Data, vol. 10, 2023
[14] J. Zheng and et al., "A hybrid framework for forecasting power generation of multiple renewable energy sources," Renewable and Sustainable Energy Reviews, vol. 172, 2023
[15] T. Hong and et al. , "Energy Forecasting: A Review and Outlook," Journals & Magazines, 2020. [Online]. Available: https://pierrepinson.com/wp-content/uploads/2020/10/Hongetal2020.pdf
[16] S. V. Chakraborty and S. K. Shukla, "Predictive Modeling of Electricity Trading Prices and the Impact of Increasing Solar Energy Penetration, , https://ieeexplore.ieee.org/document/8810536,"in 2019 IEEE Milan PowerTech Conference, Milan, 2019
[17] H. A. Alshehri, "Deep Learning for Electricity Forecasting Using Time Series Data (Theses)," 2021. [Online]. Available: https://digitalcommons.montclair.edu/etd/678
[18] A. Mohamad Elmi and et al., "Prediction of Renewable Energy Consumption of European Union Using Artificial Neural Networks," Black Sea Journal of Engineering and Science, vol. 5, no. 1, 2021
[19] D. J. Đozić and B. D. Gvozdenac Urošević, "Application of artificial neural networks for testing long-term energy policy targets," Energy, vol. 174, no. 5, 2019
[20] B. D. Gvozdenac Urošević and D. J. Đozić, "Testing long-term energy policy targets by means of artificial neural network," Energy, vol. 227, 2021
[21] D. D. Gvozdenac and et al, "Projections of Electricity Energy Mix Changes Towards Nearly Carbon Neutrality by 2050," THERMAL SCIENCE, vol. 29, no. 1, pp. 713-730, 2025, https://doiserbia.nb.rs/img/doi/0354-9836/2025/0354-98362500019G
[22] "Competitiveness of clean energy technology - CCUS," European Commission , 2023 . [Online]. Available: https://energy.ec.europa.eu/system/files/2024-01/Policy_Brief_Competitiveness_of_clean_energy_technology-CCUS.pdf
[23] "CARBON CAPTURE UTILISATION AND STORAGE IN THE EUROPEAN UNION," European Commission, 24 Oct 2023. [Online]. Available: https://setis.ec.europa.eu/carbon-capture-utilisation-and-storage-european-union-0_en
[24] "CCUS ROADMAP TO 2030," CCUS-set plan, Oct 2021. [Online]. Available: https://www.ccus-setplan.eu/wp-content/uploads/2021/11/CCUS-SET-Plan_CCUS-Roadmap-2030.pdf
[25] Y. Liu and et al., "Forecasting Shifts in Europe's Renewable and Fossil Fuel Markets Using Deep Learning Methods, Energy Science & Engineering," Energy Science & Engineering, vol. 13, p. 119-139, 2025, doi.org/10.1002/ese3.1981
[26] B. N. Oreshkin and et al., "N-BEATS neural network for mid-term electricity load forecasting," arXiv:2009.11961 [cs.LG], 2021. [Online]. Available:, 10.48550/arXiv.2009.11961
[27] A. Gasparin and et al., "Deep Learning for Time Series Forecasting: The Electric Load Case," arXiv:1907.09207 [cs.LG], 2019. [Online]. Available:, 10.48550/arXiv.1907.09207
[28] F. Kılıç, "Forecasting the Electricity Capacity and Electricity Generation Values of Wind &Solar Energy with Artificial Neural Networks Approach: The Case of Germany," Applied Artificial Intelligence, vol. 36, no. 1, 2022, doi.org/10.1080/08839514.2022.2033911
[29] M. Jovanović and et al. , "Scenarios for transitioning the electricity sector of the Republic of Serbia to sustainable climate neutrality by 2050," Utilities Policy, vol. 85, 2023, 10.1016/j.jup.2023.101681
[30] I. Goodfellow, Y. Bengio and A. Courville, Deep Learning (Adaptive Computation and Machine Learning series), MIT, 2015
[31] T. Hastie, R. Tibshirani and J. Friedman , The Elements of Statistical Learning (Second Edition), Springer, 2009
[32] S. Haykin, Neural Networks and Learning Machines, Pearson, 2008
[33] "Greenhouse gas emission intensity of electricity generation in Europe," EEA, 31 Oct 2024. [Online]. Available: https://www.eea.europa.eu/en/analysis/indicators/greenhouse-gas-emission-intensity-of-1
[34] "Life Cycle Greenhouse Gas Emissions from Electricity Generation: Update," NREL , 2021, . [Online]. Available: https://www.nrel.gov/docs/fy21osti/80580.pdf
[35] "Carbon dioxide emissions factors, Our World in Data," 2024. [Online]. Available: https://ourworldindata.org/grapher/carbon-dioxide-emissions-factor
[36] "PARIS AGREEMENT," UNFCCC, 2015. [Online]. Available: https://unfccc.int/sites/default/files/english_paris_agreement.pdf
[37] "EU Energy Roadmap 2050 (Impact Assessment and Scenario Analysis)," SEC, 15 Dec 2011. [Online]. Available: https://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=COM:2011:0885:FIN:EN:PDF
[38] "An Updated Roadmap to Net Zero Emissions by 2050," IEA, 2022. [Online]. Available: https://www.iea.org/reports/world-energy-outlook-2022/an-updated-roadmap-to-net-zero-emissions-by-2050
[39] "The European Green Deal," EUROPEAN COMMISSION, Brussels, 11 12 2019. [Online]. Available: https://eur-lex.europa.eu/resource.html?uri=cellar:b828d165-1c22-11ea-8c1f-01aa75ed71a1.0002.02/DOC_1&format=PDF
[40] A. Virone and et al., "A Vision for the EU Net-Zero Transition, CATF, April 2024, https://www.catf.us/resource/a-vision-for-the-eu-net-zero-transition/?utm_source=chatgpt.com," CATF, April 2024. [Online]. Available: https://www.catf.us/resource/a-vision-for-the-eu-net-zero-transition/?utm_source=chatgpt.com
[41] "Transformation of Europe's power system until 2050," McKinsey&Company, [Online]. Available: https://www.mckinsey.com/~/media/mckinsey/dotcom/client_service/epng/pdfs/transformation_of_europes_power_system.ashx?utm_source=chatgpt.com
[42] H. Zsiborács and et al., "Intermittent Renewable Energy Sources: The Role of Energy Storage in the European Power System of 2040," Electronics, vol. 8, no. 7, p. 729, 2019, doi.org/10.3390/electronics8070729
[43] A. Santecchia and et al. , "The critical role of electricity storage for a clean and renewable European economy," Energy and Environmental Science, 2023, 16, 5350-5370, vol. 16, pp. 5350-5370, 2023, 10.1039/D3EE02768F
PDF VERSION [DOWNLOAD]
Enhancing electricity mix forecasting with artificial neural networks: A case study of the Eu27

© 2026 Society of Thermal Engineers of Serbia. Published by the Vinča Institute of Nuclear Sciences, National Institute of the Republic of Serbia, Belgrade, Serbia. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International licence