Artificial Intelligence Assisted Solar Energy Forecasting by Explainability Approaches with LIME and SHAP

Abstract

Integrating renewable energy sources with new technologies such as artificial intelligence (AI) is important to balance energy supply and demand. The predictability of variable energy sources, such as solar energy, plays an important role in maintaining the stability and efficiency of power grids. This study examines the use of various algorithms in AI applications within renewable energy systems. The study critically evaluates existing methods and proposes an innovative approach for AI prediction in solar energy systems using advanced machine learning techniques. It focuses on the effectiveness of MLP, Ridge, and RF algorithms in forecasting Direct Current (DC). The results showed that the RF algorithm achieved the highest R² value (0.9999) and the lowest error RMSE (0.0024) and MAE (0.0006) measurements to demonstrate the superior ability of the models to explain variance in the data and make accurate predictions. In addition, the model developed with SHAP and LIME explainable AI algorithms is interpreted.

Keywords

• Solar energy
• Machine learning
• Explainable AI
• SHAP
• LIME
• Renewable energy

LIME ve SHAP ile Açıklanabilirlik Yaklaşımları ile Yapay Zeka Destekli Güneş Enerjisi Tahmini

Abstract

Yenilenebilir enerji kaynaklarını yapay zeka (AI) gibi yeni teknolojilerle entegre etmek, enerji arz ve talebini dengelemek için önemlidir. Güneş enerjisi gibi değişken enerji kaynaklarının öngörülebilirliği, elektrik şebekelerinin istikrarını ve verimliliğini korumada önemli bir rol oynamaktadır. Bu çalışma, yenilenebilir enerji sistemlerinde YZ uygulamalarında çeşitli algoritmaların kullanımını incelemektedir. Çalışma, mevcut yöntemleri eleştirel bir şekilde değerlendirmekte ve gelişmiş makine öğrenimi tekniklerini kullanarak güneş enerjisi sistemlerinde YZ tahmini için yenilikçi bir yaklaşım önermektedir. MLP, Ridge ve RF algoritmalarının Doğru Akım (DC) tahminindeki etkinliğine odaklanmaktadır. Sonuçlar, RF algoritmasının en yüksek R² değerini (0,9999) ve en düşük hata RMSE (0,0024) ve MAE (0,0006) ölçümlerini elde ederek modellerin verilerdeki varyansı açıklama ve doğru tahminler yapma konusundaki üstün yeteneğini göstermiştir. Ayrıca SHAP ve LIME açıklanabilir YZ algoritmaları ile geliştirilen model yorumlanmıştır.

Keywords

• Güneş enerjisi
• Makine öğrenimi
• Açıklanabilir YZ
• SHAP
• LIME
• Yenilenebilir enerji

References

1. [1] J. Yu, X. Li, L. Yang, L. Li, Z. Huang, K. Shen, X. Yang, X. Yang, Z. Xu, D. Zhang, and S. Du, “Deep Learning Models for PV Power Forecasting: Review,” Energies, vol. 17, no. 16, pp. 3973, 2024.
2. [2] I. Jebli, F.-Z. Belouadha, M. I. Kabbaj, and A. Tilioua, “Deep learning based models for solar energy prediction,” Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 1, pp. 349-355, 2021.
3. [3] K. R. Kumar, and M. S. Kalavathi, “Artificial intelligence based forecast models for predicting solar power generation,” Materials today: proceedings, vol. 5, no. 1, pp. 796-802, 2018.
4. [4] S. Cantillo-Luna, R. Moreno-Chuquen, D. Celeita, and G. Anders, “Deep and Machine Learning Models to Forecast Photovoltaic Power Generation,” Energies, vol. 16, no. 10, pp. 4097, 2023.
5. [5] L. M. Halabi, S. Mekhilef, and M. Hossain, “Performance evaluation of hybrid adaptive neuro-fuzzy inference system models for predicting monthly global solar radiation,” Applied Energy, vol. 213, pp. 247-261, 2018/03/01/, 2018.
6. [6] G. Zhang, X. Wang, and Z. Du, “Research on the prediction of solar energy generation based on measured environmental data,” International Journal of u-and e-Service, Science and Technology, vol. 8, no. 5, pp. 385-402, 2015.
7. [7] A. R. Kaushik, S. Padmavathi, K. S. Gurucharan, and S. C. Raja, "Performance Analysis of Regression Models in Solar PV Forecasting." pp. 1-5.
8. [8] S. Salisu, M. Mustafa, and M. Mustapha, "Predicting global solar radiation in Nigeria using adaptive neurofuzzy approach." pp. 513-521.

9. [9] B. Ersöz, M. C. Taşdelen, S. Eren, S. Sagiroglu, and A. Öter, "Solar Energy Forecasting Using Ensemble Learning Method." pp. 283-287.
10. [10] D. E. Rumelhart, G. E. Hinton, and R. J. Williams, “Learning representations by back-propagating errors,” nature, vol. 323, no. 6088, pp. 533-536, 1986.
11. [11] I. Goodfellow, Y. Bengio, and A. Courville, “Regularization for deep learning,” Deep learning, pp. 216-261, 2016.
12. [12] Y. LeCun, Y. Bengio, and G. Hinton, “Deep learning,” nature, vol. 521, no. 7553, pp. 436-444, 2015.
13. [13] R. Touza, J. Martínez Torres, M. Álvarez, and J. Roca, “Obtaining anti-missile decoy launch solution from a ship using machine learning techniques,” 2022.
14. [14] L. Breiman, “Random forests,” Machine learning, vol. 45, pp. 5-32, 2001.
15. [15] A. Liaw, “Classification and regression by randomForest,” R news, 2002.
16. [16] T. Hastie, R. Tibshirani, J. H. Friedman, and J. H. Friedman, The elements of statistical learning: data mining, inference, and prediction: Springer, 2009.
17. [17] A. E. Hoerl, and R. W. Kennard, “Ridge regression: Biased estimation for nonorthogonal problems,” Technometrics, vol. 12, no. 1, pp. 55-67, 1970.
18. [18] S. Sezer, A. Oter, B. Ersoz, C. Topcuoglu, H. İ. Bulbul, S. Sagiroglu, M. Akin, and G. Yilmaz, “Explainable artificial intelligence for LDL cholesterol prediction and classification,” Clinical Biochemistry, pp. 110791, 2024.
19. [19] S. Lundberg, “A unified approach to interpreting model predictions,” arXiv preprint arXiv:1705.07874, 2017.
20. [20] C. Molnar, Interpretable machine learning: Lulu. com, 2020.
21. [21] A. Vırıt, and A. Öter, “Kardiyovasküler Hastalıkların Derin Öğrenme Algoritmaları İle Tanısı,” Gazi Üniversitesi Fen Bilimleri Dergisi Part C: Tasarım ve Teknoloji, vol. 12, no. 4, pp. 902-912, 2024.
22. [22] C. J. Willmott, and K. Matsuura, “Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance,” Climate research, vol. 30, no. 1, pp. 79-82, 2005.
23. [23] A. C. Cameron, and F. A. Windmeijer, “An R-squared measure of goodness of fit for some common nonlinear regression models,” Journal of econometrics, vol. 77, no. 2, pp. 329-342, 1997.
24. [24] J. Neter, M. H. Kutner, C. J. Nachtsheim, and W. Wasserman, “Applied linear statistical models,” 1996.
25. [25] A. Öter, “Automatic Detection of Epileptic Seizures from EEG Signals Using Artificial Intelligence Methods,” Gazi University Journal of Science Part C: Design and Technology, pp. 1-1, 2024.
26. [26] A. Öter, B. Ersöz, Z. Berktaş, H. İ. Bülbül, E. Orhan, and Ş. Sağıroğlu, “An artificial intelligence model estimation for functionalized graphene quantum dot-based diode characteristics,” Physica Scripta, vol. 99, no. 5, pp. 056001, 2024.