SavedAdd your journal

Journals / Dokuz Eylül Üniversitesi Mühendislik Fakültesi Fen ve Mühendislik Dergisi / 2026 / Cilt: 28 Sayı: 84

Explainable Artificial Intelligence Assisted Solar Energy Forecasting: An Application with Histogram-Based Gradient Boosting Model

Açıklanabilir Yapay Zekâ Destekli Güneş Enerjisi Tahmini: Histogram Tabanlı Gradyan Artırma Modeli ile Bir Uygulama

Gökçen Özdemir(Erciyes University, Engineering Faculty, Electrical and Electronics Engineering)0000-0002-3175-6893
Pages
489–498
Publication Date
2026-09-30
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The study discusses the practical implications of using XAI-enhanced forecasting models in energy management and smart grid applications, highlighting improvements in trust, decision-making, and operational efficiency.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:The integration of solar photovoltaic (PV) energy into power grids necessitates accurate and reliable forecasting to ensure stability and efficiency.
📊
Key Findings & Data:Traditional forecasting models often lack transparency, complicating the interpretation of predictions and the optimization of energy management strategies.
💡
Conclusion & Impact:The study discusses the practical implications of using XAI-enhanced forecasting models in energy management and smart grid applications, highlighting improvements in trust, decision-making, and operational efficiency.
🔬Design:Modeling & Simulation
📍Scope:2014 · 59,112 samples
⚙️Variables:Solar PV · Güç Üretimi Tahmini · Açıklanabilir Yapay Zeka

Abstract

The integration of solar photovoltaic (PV) energy into power grids necessitates accurate and reliable forecasting to ensure stability and efficiency. Traditional forecasting models often lack transparency, complicating the interpretation of predictions and the optimization of energy management strategies. In this study, a HistGradientBoosting algorithm was employed for hourly solar power forecasting, and the SHAP (Shapley Additive Explanations) method was integrated to enhance model transparency within an Explainable Artificial Intelligence (XAI) framework. The study used the GEFCom 2014, which is an open-access dataset containing 12 different meteorological features with a total of 59,112 samples, such as solar radiation, temperature, humidity, wind speed, cloud cover, etc. The model's performance is evaluated using Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and the Coefficient of Determination. The results of the analyses show that the proposed HistGradientBoosting model achieved high prediction accuracy, with an RMSE of 0.0836 and an R² value of 0.8768. The study discusses the practical implications of using XAI-enhanced forecasting models in energy management and smart grid applications, highlighting improvements in trust, decision-making, and operational efficiency.

Özet

Güneş fotovoltaik (PV) sistemlerinin enerji şebekelerine entegrasyonu, üretim-tüketim dengesinin korunabilmesi açısından doğru ve güvenilir üretim tahminlerini zorunlu kılmaktadır. Ancak, geleneksel tahmin modellerinin yorumlanabilirlik eksikliği, bu modellerin karar verme süreçlerine duyulan güveni sınırlamaktadır. Bu çalışmada, saatlik güneş enerjisi üretimi tahmini için Histogram Tabanlı Gradyan Artırma algoritması (HistGradientBoosting) kullanılmış ve modelin şeffaflığını artırmak amacıyla Açıklanabilir Yapay Zekâ (XAI) tekniklerinden SHAP (Shapley Additive Explanations) yöntemi entegre edilmiştir. Çalışmada, açık erişimli GEFCom 2014 veri seti kullanılmış olup toplam 59.112 örneklem ve güneş ışınımı, sıcaklık, nem, bulutluluk ve rüzgâr hızı gibi 12 farklı meteorolojik öznitelik içermektedir. Model performansı, Ortalama Karesel Hata (MSE), Hatanın Karesel Ortalamasının Karekökü (RMSE), Ortalama Mutlak Hata (MAE) ve Belirleme Katsayısı (R²) metrikleriyle değerlendirilmiştir. Gerçekleştirilen analizler sonucunda, önerilen HistGradientBoosting modeli 0.0836 RMSE ve 0.8768 R² değerlerine ulaşarak yüksek bir tahmin doğruluğu sergilemiştir. SHAP analizleri, güneş ışınımının en etkili özellik olduğunu ortaya koymuştur. Çalışma, güneş enerjisi üretiminde XAI destekli tahmin modellerinin, enerji yönetimi ve akıllı şebeke uygulamalarında güvenilir karar destek sistemleri geliştirilmesine katkı sağlayabileceğini göstermektedir.

🔖

Cite This Article

Özdemir, G. (2026). Explainable Artificial Intelligence Assisted Solar Energy Forecasting: An Application with Histogram-Based Gradient Boosting Model. Dokuz Eylül Üniversitesi Mühendislik Fakültesi Fen ve Mühendislik Dergisi, Cilt: 28 Sayı: 84, 489–498. https://doi.org/10.21205/deufmd.2026288416

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (24)

  1. Abdelsattar M, Ismeil MA, Zayed MMAA, Abdelmoety A, Emad-Eldeen A. Assessing Machine Learning Approaches for Photovoltaic Energy Prediction in Sustainable Energy Systems. IEEE Access 2024;12:107599-107615. doi:10.1109/Access.2024.3437191.
  2. Yu YJ, Cao JF, Zhu JY. An LSTM Short-Term Solar Irradiance Forecasting Under Complicated Weather Conditions. IEEE Access 2019;7:145651-145666. doi:10.1109/Access.2019.2946057.
  3. Gaboitaolelwe J, Zungeru AM, Yahya A, Lebekwe CK, Vinod DN, Salau AO. Machine Learning Based Solar Photovoltaic Power Forecasting: A Review and Comparison. IEEE Access 2023;11:40820-40845. doi:10.1109/Access.2023.3270041.
  4. Dimd BD, Voller S, Cali U, Midtgard OM. A Review of Machine Learning-Based Photovoltaic Output Power Forecasting: Nordic Context. IEEE Access 2022;10:26404-26425. doi:10.1109/Access.2022.3156942.
  5. Castillo-Rojas W, Salinas JP. Forecasting Models Applied in Solar Photovoltaic and Wind Energy: A Systematic Mapping Study. IEEE Access 2024;12:151092-151111. doi:10.1109/Access.2024.3471073.
  6. Yagli GM, Yang DZ, Srinivasan D. Automatic hourly solar forecasting using machine learning models. Renew Sust Energ Rev 2019;105:487-498. doi:10.1016/j.rser.2019.02.006.
  7. Ledmaoui Y, El Maghraoui A, El Aroussi M, Saadane R, Chebak A, Chehri A. Forecasting solar energy production: A comparative study of machine learning algorithms. Energy Rep 2023;10:1004-1012. doi:10.1016/j.egyr.2023.07.042.
  8. Arrieta AB, vd. Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Inform Fusion 2020;58:82-115. doi:10.1016/j.inffus.2019.12.012.
  9. DARPA Agency. XAI: Explainable Artificial Intelligence. Erişim adresi: https://www.darpa.mil/research/programs/explainable-artificial-intelligence [Erişim tarihi 12 Haziran 2025].
  10. Saraswat D, vd. Explainable AI for Healthcare 5.0: Opportunities and Challenges. IEEE Access 2022;10:84486-84517. doi:10.1109/Access.2022.3197671.
  11. Capuano N, Fenza G, Loia V, Stanzione C. Explainable Artificial Intelligence in CyberSecurity: A Survey. IEEE Access 2022;10:93575-93600. doi:10.1109/Access.2022.3204171.
  12. Richmond KM, Muddamsetty SM, Gammeltoft-Hansen T, Olsen HP, Moeslund TB. Explainable AI and law: An evidential survey. Digital Society 2024;3(1):1.
  13. Ozdemir G, Kuzlu M, Sarp S, Catak FO, Dimd BD, Cali U. The role of eXplainable Artificial Intelligence (XAI) in smart grids. İçinde: Big data application in power systems. Elsevier; 2024, s. 349-370.
  14. Kuzlu M, Cali U, Sharma V, Güler Ö. Gaining insight into solar photovoltaic power generation forecasting utilizing explainable artificial intelligence tools. IEEE Access 2020;8:187814-187823.
  15. Ozdemir G, Kuzlu M, Catak FO. Machine learning insights into forecasting solar power generation with explainable AI. Electrical Engineering 2024:1-22.
  16. Hong T, Pinson P, Fan S, Zareipour H, Troccoli A, Hyndman RJ. Probabilistic energy forecasting: Global Energy Forecasting Competition 2014 and beyond. International Journal of Forecasting 2016;32(3):896-913.
  17. Ke G, Meng Q, Finley T, Wang T, Chen W, Ma W, vd. LightGBM: A Highly Efficient Gradient Boosting Decision Tree. Advances in Neural Information Processing Systems 2017;30:3146-3154.
  18. Scikit-learn Developers. Histogram-based Gradient Boosting. 2019. Erişim adresi: https://scikit-learn.org/stable/modules/ensemble.html#histogram-based-gradient-boosting
  19. Shapley L. A Value for n-Person Games. In: Kuhn, H. and Tucker, A., Eds., Contributions to the Theory of Games II, Princeton University Press, Princeton, 1953.
  20. Lundberg SM, Lee SI. A Unified Approach to Interpreting Model Predictions. Advances in Neural Information Processing Systems (NIPS) 2017;30.
  21. Singh U, Singh S, Gupta S, Alotaibi MA, Malik H. Forecasting rooftop photovoltaic solar power using machine learning techniques. Energy Reports 2025;11. doi:10.1016/j.egyr.2025.03.005.
  22. Parczyk P, Burduk R. Short- and medium-term horizon photovoltaic power forecasting based on terrestrial solar radiation features. Engineering Applications of Artificial Intelligence 2025;139:112668. doi:10.1016/j.engappai.2025.112668.
  23. Erbay C. Machine learning models for solar forecasting and impact on green hydrogen production costs. International Journal of Hydrogen Energy 2025. doi:10.1016/j.ijhydene.2025.04.465.
  24. Hayawi K, Maliakkal H, Venugopal N, Hussain TM, Rajagopalan GB. Machine learning - driven solar forecasting in dust-prone regions for sustainable energy systems. Sustainable Energy Journal of Africa 2025;10:100108. doi:10.1016/j.seja.2025.100108.