Comparison of The Performance of Machine Learning Methods for Solar Energy Power Prediction

Year 2026, Volume: 15 Issue: 2, 189 - 195, 29.01.2026

Özlem Bezek Güre

Abstract

In parallel with population growth, the demand for electrical energy continues to rise. As traditional energy sources have proved insufficient to meet this demand, attention has shifted to environmentally friendly renewable energy sources, among which solar energy stands out. Electricity generation from solar energy is achieved via photovoltaic (PV) systems. Accurately predicting generated power using machine learning (ML) methods with low error supports effective and efficient electricity planning. This study aims to predict solar energy power using ML methods—Gradient Boosting Regressor (GBR), Decision Tree Regressor (DTR), Random Forest Regressor (RFR), and AdaBoost Regressor (AR). For this purpose, the “Solar energy power generation” dataset from Kaggle was utilized. The dataset includes meteorological variables such as temperature, humidity, and radiation, consisting of 21 variables and 4,212 measurements. The Relief feature selection method was used to identify the most informative independent variables for predicting solar energy power. Seventy percent of the data was used for training and the remainder for testing. Model performance was examined with respect to the number of variables. To compare methods, we employed MSE, RMSE, MAE, R2, and training time, and applied cross-validation to enhance model performance. The results indicate that AR achieved superior predictive performance, yielding lower errors (MSE = 0.111, RMSE = 0.333, MAE = 0.145) than the other methods. In contrast, RFR was found to be better in terms of both speed (training time = 2.896) and explained variance (R2=0.926). Additionally, except for DTR, all evaluated techniques exhibited improved performance as the number of variables increased.

Keywords

Machine Learning , Predictive Models , Number of Variables , Regression , Solar Energy , Power Generation

GÜNEŞ ENERJİSİ GÜÇ TAHMİNİ İÇİN MAKİNE ÖĞRENME YÖNTEMLERİNİN PERFORMANSLARININ KARŞILAŞTIRILMASI

Abstract

Bu çalışmada, makine öğrenme yöntemlerinden Gradient Boosting Regressor (GBR), Decision Tree Regressor (DTR), Random Forest Regressor (RFR) ve Adaboost Regressor (AR) kullanılarak solar enerji gücünü tahminlemek amaçlanmaktadır. Bu amaçla, Kaggle veri tabanında yer alan “Solar energy power generation dataset” kullanılmıştır. Veri seti 21 değişken 4212 adet ölçümden oluşmaktadır. Çalışmada Relief özellik seçim yöntemi kullanılarak en önemli görülen 4 bağımsız değişken güneş enerji gücünü tahminlemek amacıyla kullanılmıştır. Verilerin %70’i eğitim geriye kalanları test verisi olarak ayrılmıştır. Değişken sayılarına göre yöntemlerin performansı incelenmiştir. Yöntemlerin performanslarını karşılaştırmak amacıyla MSE, RMSE, MAE, R2 ve eğitim zamanı ölçülerinden yararlanılmıştır. Daha iyi model performans elde edebilmek için çapraz doğrulama yöntemi kullanılmıştır. Analiz sonuçlarına göre GBR yönteminin diğer yöntemlere nazaran daha düşük hatalar ile daha başarılı tahminleme performansı gösterdiği belirlenmiştir. Diğer taraftan; hız ve R2 ölçüsü açısından ise RFR yönteminin daha iyi olduğu görülmektedir. Çalışmada DT yönteminin dışındaki tüm yöntemlerde, değişken sayısı arttıkça yöntemlerin daha iyi performans gösterdiği belirlenmiştir.

Keywords

MAkine öğrenme , Tahmin modelleri , Değişken sayısı , Regresyon , Güneş enerjisi

References

• [1] Panwar, N, L,, Kaushik, S, C,, & Kothari, S, (2011), Role of renewable energy sources in environmental protection: A review, Renewable and sustainable energy reviews, 15(3), 1513-1524,
• [2] Tasnin, W,, Saikia, L, C,, & Raju, M, (2018), Deregulated AGC of multi-area system incorporating dish-Stirling solar thermal and geothermal power plants using fractional order cascade controller, International Journal of Electrical Power & Energy Systems, 101, 60-74,
• [3] Borunda, M,, Ramírez, A,, Garduno, R,, Ruíz, G,, Hernandez, S,, & Jaramillo, O, A, (2022), Photovoltaic power generation forecasting for regional assessment using machine learning, Energies, 15(23), 8895,
• [4] Nur, A,, Güre, B,, Rüstemli, S, & Bezek Güre, Ö, (2024) Solar Power Estimation by Using Artificial Neural Networks, he International Conference on Energy and Environmental Technologies in Engineering and Architecture (ICETEA 2024,
• [5] Deng, X,, Da, F,, Shao, H,, & Wang, X, (2023), A Survey of the Researches on Grid-Connected Solar Power Generation Systems and Power Forecasting Methods Based on Ground-Based Cloud Atlas, Energy Engineering, 120(2), 385-408,
• [6] Shahid, F,, Zameer, A,, Afzal, M,, & Hassan, M, (2020), Short term solar energy prediction by machine learning algorithms, arXiv preprint arXiv:2012,00688,
• [7] Nourani, V,, Elkiran, G,, Abdullahi, J,, & Tahsin, A, (2019), Multi-region modeling of daily global solar radiation with artificial intelligence ensemble, Natural Resources Research, 28, 1217-1238,
• [8] Ateş, K, T, (2022), Çok Katmanlı Yapay Sinir Ağı Modeli ve Kültürel Algoritma Modeli Kullanılarak Geliştirilen Melez Yöntem ile Kısa Vadeli Fotovoltaik Enerji Santrali Çıkış Gücü Tahmini, Osmaniye Korkut Ata Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 5(1), 342-354,
• [9] Yang, H,-T,; Huang, C,-M,; Huang, Y,-C,; Pai, Y,-S, A weather-based hybrid method for 1-day ahead hourly forecasting of solar power output, IEEE Trans, Sustain, Energy 2014, 5, 917–926,
• [10] Abdullah, B, U, D,, Khanday, S, A,, Islam, N, U,, Lata, S,, Fatima, H,, & Nengroo, S, H, (2024), Comparative Analysis Using Multiple Regression Models for Forecasting Photovoltaic Power Generation, Energies, 17(7), 1564,
• [11] Sarker, I, H, (2021), Machine learning: Algorithms, real-world applications and research directions, SN computer science, 2(3), 160,
• [12] Nasteski, V, (2017), An overview of the supervised machine learning methods, Horizons, b, 4(51-62), 56,
• [13] Rajulu, A, G,, Mohanty, P,, Pati, U, C,, & Mahapatra, K, (2024, November), SolarPred: Hybrid Machine Learning Based Time Series Prediction Model for Solar Energy Applications, In 2024 IEEE 4th International Conference on Applied Electromagnetics, Signal Processing, & Communication (AESPC) (pp, 1-6), IEEE,
• [14] Sandeep, C, S,, Mohanty, P,, & Pati, U, C, (2023, August), Machine learning based framework for prediction of photovoltaic output power, In 2023 IEEE 3rd International Conference on Sustainable Energy and Future Electric Transportation (SEFET) (pp, 1-6), IEEE,
• [15] Erten, M, Y,, & Aydilek, H, (2022), Solar power prediction using regression models, International Journal of Engineering Research and Development, 14(3), 333-342,
• [16] Babu, A. R. V., Kumar, N. B., Narasipuram, R. P., Periyannan, S., Hosseinpour, A., & Flah, A. (2025). Solar Energy Forecasting using Machine Learning Techniques for Enhanced Grid Stability. IEEE Access.
• [17] Büyükbiçakci, E,, Atli, C,, Dumanli, M,, & Bulat, S, (2024, December), Enhanced Photovoltaic Power Forecasting in Renewable Energy Systems Using Genetic Algorithm and SVM Approaches, In 2024 4th International Conference on Mobile Networks and Wireless Communications (ICMNWC) (pp, 1-6), IEEE,
• [18] Mohanty, P,, Subhadarshini, K,, Nayak, R,, Pati, U, C,, & Mahapatra, K, (2025), Exploring data-driven multivariate statistical models for the prediction of solar energy, In Computer Vision and Machine Intelligence for Renewable Energy Systems (pp, 85-101), Elsevier,
• [19] Pachauri, R, K,, Yadav, J,, Torto, S, O, G,, Minai, A, F,, Pandey, V,, Shashikant, & Sharma, P, (2024), Solar Photovoltaic System‐Based Power Generation: Imperative Role of Artificial Intelligence and Machine Learning, Clean and Renewable Energy Production, 267-286,
• [20] Chang, J,, & Jia, S, Y, (2009, July), Modeling and application of wind-solar energy hybrid power generation system based on multi-agent technology, In 2009 International Conference on Machine Learning and Cybernetics (Vol, 3, pp, 1754-1758), IEEE,
• [21] Mohanty, P,, Pati, U, C,, & Mahapatra, K, (2023, October), Deep Learning Based Framework for Forecasting Solar Panel Output Power, In IFIP International Internet of Things Conference (pp, 229-239), Cham: Springer Nature Switzerland,
• [22] https://www,kaggle,com/datasets/stucom/solar-energy-power-generation-dataset
• [23] Huang, Y,, Liu, Y,, Li, C,, & Wang, C, (2019), GBRTVis: online analysis of gradient boosting regression tree, Journal of Visualization, 22, 125-140,
• [24] Gumaei, A,, Al-Rakhami, M,, Al Rahhal, M, M,, Albogamy, F, R,, Al Maghayreh, E,, & AlSalman, H, (2021), Prediction of COVID-19 confirmed cases using gradient boosting regression method, Comput Mater Continua, 66, 315-329,
• [25] Touzani, S,, Granderson, J,, & Fernandes, S, (2018), Gradient boosting machine for modeling the energy consumption of commercial buildings, Energy and Buildings, 158, 1533-1543,
• [26] Zhang, Y,, & Haghani, A, (2015), A gradient boosting method to improve travel time prediction, Transportation Research Part C: Emerging Technologies, 58, 308-324,
• [27] Islam, S,, & Amin, S, H, (2020), Prediction of probable backorder scenarios in the supply chain using Distributed Random Forest and Gradient Boosting Machine learning techniques, Journal of Big Data, 7(1), 65,
• [28] Pathak, S,, Mishra, I,, & Swetapadma, A, (2018, November), An assessment of decision tree based classification and regression algorithms, In 2018 3rd International Conference on Inventive Computation Technologies (ICICT) (pp, 92-95), IEEE,
• [29] Chang, C, L,, & Chen, C, H, (2009), Applying decision tree and neural network to increase quality of dermatologic diagnosis, Expert Systems with Applications, 36(2), 4035-4041,
• [30] Chen, M, Y, (2011), Predicting corporate financial distress based on integration of decision tree classification and logistic regression, Expert systems with applications, 38(9), 11261-11272,
• [31] Pekel, E, (2020), Estimation of soil moisture using decision tree regression, Theoretical and Applied Climatology, 139(3-4), 1111-1119,
• [32] Patil, S,, Patil, A,, & Phalle, V, M, (2018, December), Life prediction of bearing by using adaboost regressor, In Proceedings of TRIBOINDIA-2018 An International Conference on Tribology,
• [33] Xu, M,, Watanachaturaporn, P,, Varshney, P, K,, & Arora, M, K, (2005), Decision tree regression for soft classification of remote sensing data, Remote Sensing of Environment, 97(3), 322-336,
• [34] Wang, L,a,, Zhou, X,, Zhu, X,, Dong, Z,, & Guo, W, (2016), Estimation of biomass in wheat using random forest regression algorithm and remote sensing data, The Crop Journal, 4, 212-219,
• [35] Biau, G,, Cadre, B,, & Rouvìère, L, (2019), Accelerated gradient boosting, Machine learning, 108, 971-992,
• [36] Güre, Ö, B,, Kayri, M,, & Erdoğan, F, (2020), Analysis of Factors Effecting PISA 2015 Mathematics Literacy via Educational Data Mining, Education & Science/Egitim ve Bilim, 45(202),
• [37] Güre, Ö, B,, Şevgin, H,, & Kayri, M, (2023), Reviewing the Factors Affecting PISA Reading Skills by Using Random Forest and MARS Methods, International Journal of Contemporary Educational Research, 10(1), 181196,
• [38] Shanmugasundar, G,, Vanitha, M,, Čep, R,, Kumar, V,, Kalita, K,, & Ramachandran, M, (2021), A comparative study of linear, random forest and adaboost regressions for modeling non-traditional machining, Processes, 9(11), 2015,
• [39] Li, Y,, Zou, C,, Berecibar, M,, Nanini-Maury, E,, Chan, J, C, W,, Van den Bossche, P,, ,,, & Omar, N, (2018), Random forest regression for online capacity estimation of lithium-ion batteries, Applied energy, 232, 197-210,
• [40] Geneur, R,, Poggi, J, M,, Tuleao Malot, C, ve Villa-Vialaneix, N, (2017), Random forest for big data, Big Data Research
• [41] Bezek Güre, Ö, (2023), Investigation of ensemble methods in terms of statistics: TIMMS 2019 example, Neural Computing and Applications, 35(32), 23507-23520
• [42] Ferreira, A, J,, & Figueiredo, M, A, (2012), Boosting algorithms: A review of methods, theory,
• [43] Ying, C,, Qi-Guang, M,, Jia-Chen, L,, & Lin, G, (2013), Advance and prospects of AdaBoost algorithm, Acta Automatica Sinica, 39(6), 745-758,
• [44] Shrestha, D, L,, & Solomatine, D, P, (2006), Experiments with AdaBoost, RT, an improved boosting scheme for regression, Neural computation, 18(7), 1678-1710,
• [45] Gupta, K, K,, Kalita, K,, Ghadai, R, K,, Ramachandran, M,, & Gao, X, Z, (2021), Machine learning-based predictive modelling of biodiesel production—A comparative perspective, Energies, 14(4), 1122,
• [46] Kummer, N,, & Najjaran, H, (2014), Adaboost, MRT: Boosting regression for multivariate estimation, Artif, Intell, Res,, 3(4), 64-76,
• [47] Wang, F,, Li, Z,, He, F,, Wang, R,, Yu, W,, & Nie, F, (2019), Feature learning viewpoint of AdaBoost and a new algorithm, IEEE Access, 7, 149890-149899,
• [48] Rokach L (2010) Ensemble-based classifiers, Artif Intell Rev 33(1):1–39
• [49] Sagi O, Rokach L (2018) Ensemble learning: a survey, Wiley Interdiscip Rev Data Min Knowl Discov 8(4):e1249
• [50] B. Kégl. Introduction to AdaBoost. Accessed: Nov. 15, 2009. [Online]. Available: https://users.lal.in2p3.fr/kegl/teaching/stages/notes/tutorial
• [51] Kayri, M, (2015), An intelligent approach to educational data: Performance comparison of the multilayer perceptron and the radial basis function artificial neural networks, Educational Sciences: Theory & Practice, 15(5),
• [52] Kayri, M,, Kayri, I,,& Gençoğlu, M, T, (2017), The performance comparison of multiple linear regression, random forest and artificial neural network by using photovoltaic and atmospheric data, In 2017 14th International Conference on Engineering of Modern Electric Systems (EMES), 1–4,
• [53] Keller, P, S,, El-Sheikh, M,, Granger, D, A,, & Buckhalt, J, A, (2012), Interactions between salivary cortisol and alphaamylase as predictors of children’s cognitive functioning and academic performance, Physiology & Behavior, 105, 987-995,
• [54] Le, T. T., Urbanowicz, R. J., Moore, J. H., & McKinney, B. A. (2019). Statistical inference relief (STIR) feature selection. Bioinformatics, 35(8), 1358-1365.
• [55] Sordo M, Zeng Q (2005) On sample size and classification accuracy: a performance comparison. In: International symposium on biological and medical data analysis. Springer, Berlin, pp 193–201
• [56] Temel GO, Erdog˘an S, Ankaralı H (2012) Usage of resampling methods for evaluating the performance of classification model. J Inf Technol 5(3):1–8
• [57] Dolgun MO¨ (2014) Comparison of the performances of data mining classification methods based on prevalence of the dependent variable, sample size and the correlation of the independent variables. PhD thesis, Hacettepe University
• [58] Kanik EA, Temel GO, Erdog˘an S, Kaya I˙E (2013) Affected states soft independent modeling by class analogy from the relation between independent variables, number of independent variables and sample size. Balkan Med J 2013(1):28–32
• [59] Yabacı A (2017) Comparison of tree-based methods used in survival data. PhD thesis, Uludag˘ University
• [60] Nuray SE, Genc¸dal HB, Arama ZA (2021) Zeminlerin kıvam ve kompaksiyon o¨zelliklerinin tahmininde rastgele orman regresyonu yo¨nteminin uygulanabilirlig˘i. Mu¨hendislik Bilimleri ve Tasarım Dergisi 9(1):265–281
• [61] Kumar S, Chong I (2018) Correlation analysis to identify the effective data in machine learning: prediction of depressive disorder and emotion states. Int J Environ Res Public Health 15(12):2907
• [62] Latha CBC, Jeeva SC (2019) Improving the accuracy of prediction of heart disease risk based on ensemble classification techniques. Inf Med Unlock 16:100203
• [63] Sun Y, Li Z, Li X, Zhang J (2021) Classifier selection and ensemble model for multi-class imbalance learning in education grants prediction. Appl Artif Intell 35(4):290–303
• [64] Ceyhan G (2020) Comparison of performance of data mining methods used for classification in terms of data characteristics. PhD thesis, Gazi University
• [65] Kwon O, Sim JM (2013) Effects of data set features on the performances of classification algorithms. Expert Syst Appl 40(5):1847–1857
• [66] Dag˘ H, Sayın KE, Yenidog˘an I, Albayrak S, Acar C (2012) Comparison of feature selection algorithms for medical data. In: 2012 International symposium on innovations in intelligent systems and applications. IEEE, pp 1–5
• [67] Huang S, Fang N (2013) Predicting student academic performance in an engineering dynamics course: a comparison of four types of predictive mathematical models. Comput Educ 61:133–145
• [68] Kasap Y, Dog˘an N, Koc¸ak C (2021) Determining variables that predict reading comprehension success by data mining in PISA 2018. Manisa Celal Bayar Univ J Soc Sci 19(4):241–258
• [69] Bauer E, Kohavi R (1999) An empirical comparison of voting classification algorithms: bagging boosting and variants. Mach Learn 36(1):105–139
• [70] Rokach L (2010) Ensemble-based classifiers. Artif Intell Rev 33(1):1–39
• [71] Dimitriadou E, Weingessel A, Hornik K (2003) A cluster ensembles framework. Design and application of hybrid intelligent systems. IOS Press, Amsterdam
• [72] Sidhu, R. K., Kumar, R., & Rana, P. S. (2020). Machine learning based crop water demand forecasting using minimum climatological data. Multimedia Tools and Applications, 79(19), 13109-13124.
• [73] Geçmez, A,, & Gençer, Ç, (2020), Güneş Enerji Santrali Üretim Verilerinin Meteorolojik Verilere Bağlı olarak Yapay Zeka Yöntemleri ile Tahmini, Doç, Dr, Aydın Ruşen Doç, Dr, Sadık Alper Yıldızel, 39,
• [74] Uğuz, S,, Oral, O,, & Çağlayan, N, (2019), PV Güç Santrallerinden Elde Edilecek Enerjinin Makine Öğrenmesi Metotları Kullanılarak Tahmin Edilmesi, International Journal of Engineering Research and Development, 11(3), 769-779,
• [75] Guyon, I., & Elisseeff, A. (2003). An introduction to variable and feature selection. Journal of Machine Learning Research, 3, 1157–1182.