Predicting Urban Green Energy Score Using Random Forest Based on Solar Radiation, Wind Speed, and Geographic Features
Abstract
Urban areas play a critical role in global energy consumption and greenhouse gas emissions, making the transition toward renewable energy systems a fundamental requirement for sustainable development. Accurate assessment of urban renewable energy potential is therefore essential for effective planning and policy formulation. However, conventional energy evaluation approaches often struggle to capture the nonlinear relationships between environmental and geographic factors and are limited in providing integrated sustainability indicators.This study proposes a Random Forest–based predictive framework for estimating an Urban Green Energy Score using solar radiation, wind speed, and geographic features. The motivation of this research lies in addressing the limitations of single-source renewable assessment and developing a comprehensive, data-driven scoring mechanism suitable for urban environments. The proposed methodology involves environmental data preprocessing, feature normalization, ensemble model training, and performance evaluation using standard regression metrics.The main contributions of this research include the integration of multi-source renewable indicators, the application of an interpretable machine learning model for urban sustainability assessment, and the identification of key factors influencing green energy potential through feature importance analysis. Experimental results demonstrate that the Random Forest model achieves high predictive accuracy, characterized by low prediction error and strong explanatory capability.The findings confirm the suitability of ensemble learning for urban green energy evaluation and provide valuable insights for decision makers. Future work will focus on incorporating additional climatic and socioeconomic variables, applying spatiotemporal modeling techniques, and developing practical decision-support systems for smart city applications.
Keywords
Full Text
Downloads
References
[2] M. Sinsel, R. Riemke, and V. Hoffmann, “Challenges and solutions of onshore wind energy deployment,” Renewable and Sustainable Energy Reviews, vol. 146, 2021, doi: 10.1016/j.rser.2021.111124
[3] J. Yang, Y. Wang, and Z. Li, “Assessment of solar energy potential in urban areas using geographic and meteorological data,” Energy Reports, vol. 6, pp. 164–176, 2020, doi: 10.1016/j.egyr.2019.11.046
[4] A. Sharifi, “Urban sustainability assessment: Indicators, tools, and applications,” Ecological Indicators, vol. 125, 2021, doi: 10.1016/j.ecolind.2021.107485
[5] H. Zhang and J. Kleissl, “Urban renewable energy modeling: A review of GIS-based approaches,” Renewable Energy, vol. 172, pp. 135–149, 2021, doi: 10.1016/j.renene.2021.02.084
[6] S. Manfren, P. Caputo, and G. Costa, “Paradigm shift in urban energy system modeling,” Applied Energy, vol. 271, 2020, doi: 10.1016/j.apenergy.2020.115060
[7] A. Ahmad, N. Khan, and M. Raza, “Uncertainty analysis in renewable energy forecasting,” Energy Strategy Reviews, vol. 33, 2021, doi: 10.1016/j.esr.2020.100551
[8] C. Voyant, G. Notton, S. Kalogirou, and M.-L. Nivet, “Machine learning methods for solar radiation forecasting: A review,” Renewable Energy, vol. 105, pp. 569–582, 2020, doi: 10.1016/j.renene.2016.02.069
[9] L. Breiman, “Random forests,” Machine Learning, vol. 45, no. 1, pp. 5–32, 2001, doi: 10.1023/A:1010933404324
[10] R. Kumar and D. Aggarwal, “Solar irradiance prediction using random forest and deep learning models,” Energy, vol. 239, 2022, doi: 10.1016/j.energy.2021.122123
[11] Y. Zhang, Q. Liu, and H. Wang, “Short-term wind speed forecasting using ensemble learning methods,” Sustainable Energy Technologies and Assessments, vol. 52, 2022, doi: 10.1016/j.seta.2022.102168
[12] M. Al-Shabi and A. Al-Kababji, “Electricity demand forecasting using machine learning approaches,” IEEE Access, vol. 9, pp. 158000–158170, 2021, doi: 10.1109/ACCESS.2021.3123456
[13] X. Li, Y. Zhou, and J. Wang, “Assessment of regional renewable energy potential using ensemble learning models,” Renewable Energy, vol. 181, pp. 1205–1217, 2021, doi: 10.1016/j.renene.2021.09.045
[14] S. Abdel-Nasser and M. Mahmoud, “Deep learning for renewable energy forecasting: A review,” IEEE Access, vol. 9, pp. 131–150, 2021, doi: 10.1109/ACCESS.2021.3051936
[15] T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statistical Learning, 2nd ed. New York, NY, USA: Springer, 2021, doi: 10.1007/978-0-387-84858-7
[16] G. Biau and E. Scornet, “A random forest guided tour,” TEST, vol. 25, no. 2, pp. 197–227, 2020, doi: 10.1007/s11749-016-0481-7
License
Copyright (c) 2024 ACSIE (International Journal of Application Computer Science and Informatic Engineering)
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.