Solar Radiation Computation from Satellite Weather Data in Batam Using Linear Regression, Random Forest, and Decision Tree
Main Article Content
Abstract
This study addresses the necessity of evaluating solar radiation as a renewable energy source in tropical regions, specifically focusing on the challenges of estimation in Batam. The objective is to model daily solar radiation levels using satellite-derived weather data to overcome the lack of surface observation stations. Daily meteorological variables, including air temperature, relative humidity, rainfall, surface pressure, and wind speed, were sourced from the NASA POWER platform for the period January 1, 2020, to July 2, 2025. To ensure robust model generalization and prevent data leakage, the dataset was partitioned chronologically, utilizing data from 2020–2024 for training and the year 2025 for independent testing. Three computational models Linear Regression (LR), Random Forest (RF), and Decision Tree (DT) were applied to the processed data. The evaluation results indicate that the Random Forest model achieved the highest relative performance among the tested algorithms, recording a Mean Squared Error (MSE) of 19.61, a Mean Absolute Error (MAE) of 3.42, and a coefficient of determination R² of 0.20. In comparison, the Linear Regression model produced an R² of 0.19, while the Decision Tree showed significantly lower predictive accuracy. Despite being the most viable model, an R² of 0.20 reveals that the current predictors explain only 20% of the variance in solar radiation, highlighting the inherent complexity of tropical atmospheric dynamics. These findings suggest that while machine learning offers a promising framework for energy planning in Batam, further research incorporating additional explanatory features, such as cloud cover or aerosol indices, is required to improve model reliability.
Downloads
References
[1] L. Huang, J. Kang, M. Wan, L. Fang, C. Zhang, and Z. Zeng, “Prediction of Solar Radiation Using Various Machine Learning Algorithms and Its Implications for Extreme Climate Events,” Front Earth Sci (Lausanne), vol. 9, April 2021, doi: 10.3389/feart.2021.596860.
[2] J. Fan et al., “Empirical Models and Machine Learning for Predicting Daily Global Solar Radiation from Sunshine Duration: A Review and Case Study in China,” Renewable and Sustainable Energy Reviews, vol. 100, pp. 186–212, Feb. 2019, doi: 10.1016/j.rser.2018.10.018.
[3] M. Jamei et al., “Data-Based Model for Predicting Solar Radiation in Semi-Arid Regions,” Computers, Materials and Continua, vol. 74, no. 1, pp. 1625–1640, 2023, doi: 10.32604/cmc.2023.031406.
[4] R. Ismayanti and W. Maulana Baihaqi, “Predicting the Potential of an Area to Become a Solar Power Plant Using Machine Learning,” JIITE.
[5] V. Demir, “Evaluation of Solar Radiation Prediction Models Using AI: Performance Comparison in the High-Potential Region of Konya, Türkiye,” Atmosphere (Basel), vol. 16, no. 4, Apr. 2025, doi: 10.3390/atmos16040398.
[6] Q. Li, M. Bessafi, and P. Li, “Mapping Surface Solar Radiation Predictions with a Linear Regression Model: A Case Study on Reunion Island,” Atmosphere (Basel), vol. 14, no. 9, September 2023, doi: 10.3390/atmos14091331.
[7] L. P. Darman, Januhariadi, M. P. Yudha, and Aslan, “Assessment of NASA POWER Reanalysis Products as an Alternative Data Source for Weather Monitoring in West Sumbawa, Indonesia,” in E3S Web of Conferences, EDP Sciences, Feb. 2024. doi: 10.1051/e3sconf/202448506006.
[8] I. K. Tanoli et al., “Machine learning for high-performance solar radiation prediction,” Energy Reports, vol. 12, pp. 4794–4804, Dec. 2024, doi: 10.1016/j.egyr.2024.10.033.
[9] P. Setiawati et al., “SOLAR ENERGY GENERATION PREDICTION: A Machine Learning Approach for Network Stability and Efficiency,” Jurnal Pilar Nusa Mandiri, vol. 21, no. 1, pp. 34–43, Mar. 2025, doi: 10.33480/pilar.v21i1.6126.
[10] U. V. Teja, M. Sai Kiran, V. Karthikeya, M. E. Murali, and T. Kumanan, “Prediction of Solar Radiation Using Machine Learning and Python.”
[11] R. Srivastava, A. N. Tiwari, and V. K. Giri, “Prediction of Solar Radiation Using MARS, CART, M5, and Random Forest Models: A Case Study for India,” Heliyon, vol. 5, no. 10, October 2019, doi: 10.1016/j.heliyon.2019.e02692.
[12] D. Ardiansyah, “COMPARISON OF SUNRADIATION PREDICTION MODELS BASED ON MACHINE LEARNING AT THE FATMAWATI SOEKARNO BENGKULU METEOROLOGICAL STATION,” Megasains, vol. 14, no. 1, Sep. 2023, doi: 10.46824/megasains.v14i1.129.
[13] C. G. Villegas-Mier, J. Rodriguez-Resendiz, J. M. Álvarez-Alvarado, H. Jiménez-Hernández, and Á. Odry, “Optimized Random Forest for Solar Radiation Prediction Using Sunlight Hours,” Micromachines (Basel), vol. 13, no. 9, September 2022, doi: 10.3390/mi13091406.
[14] M. Attya, O. Abo-Seida, H. Mohamed, and A. Mohammed, “A Hybrid Deep Learning Framework for Solar Radiation Prediction Based on Satellite Images and Regional Data,” Neural Comput Appl, July 2025, doi: 10.1007/s00521-025-11197-3.
[15] A. N. Oktaviani et al., “Analysis of Solar Radiation Prediction Using Machine Learning Algorithms and Bayesian Optimization Implementation in the Province of DKI Jakarta,” Journal of Information Management & Information Systems (MISI), vol. 8, no. 1, 2025, doi: 10.36595/misi.v5i2.
[16] A. Mujtaba, “Towards Sustainable Energy Policies: Machine Learning Applications in Projecting Bio Solar Consumption in Indonesia,” 2024, doi: 10.18196/jsp/v15i1.360.
Article Sidebar
