Hydroelectric Power Forecasting via XGBoost (Extreme Gradient Boosted Decision Trees)

Öz

Hydropower energy is crucial for meeting Turkey's growing energy demand, driven by rapid economic and population growth. Its seasonal variability makes it suitable for forecasting algorithms. This study aims to predict energy production at the EÜAŞ Aslantaş Hydroelectric Power Plant, with a capacity exceeding 100 MW. A forecasting model was created using XGBoost (Extreme Gradient Boosted Decision Trees) incorporating inputs such as timestamp, historical energy production data, and temperature. Energy data source is the EPİAŞ Transparency Platform and the data was processed with Python. The model was optimized by varying the number of trees and learning rates (η). Its effectiveness was rigorously assessed using statistical metrics, including the coefficient of determination (R²), Mean Absolute Scaled Error (MASE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), Weighted Absolute Percentage Error (WAPE). The results indicate that machine learning algorithms can significantly enhance hydropower energy forecasting and optimize energy management strategies.

Anahtar Kelimeler

• HPP
• Hydroelectricity
• Forecasting
• XGBoost

XGBoost (Aşırı Gradyan Artırımlı Karar Ağaçları) ile Hidroelektrik Enerji Tahmini

Öz

Hidroelektrik enerji, Türkiye'nin hızlı ekonomik ve nüfus artışıyla artan enerji talebinin karşılanmasında büyük önem taşır. Mevsimsel bağımlılığı nedeniyle hidroelektrik enerji, tahmin algoritmaları için uygundur. Bu çalışma, Türkiye'de 100 MW'ın üzerinde güç üreten EÜAŞ Aslantaş HES'de enerji üretimini tahmin etmeyi amaçlamaktadır. Tahmin modeli, XGBoost (Aşırı Gradyan Artırımlı karar ağaçları) ile tarih-saat kayıtları, geçmiş enerji üretim verileri ve sıcaklık gibi çeşitli girdi kullanılarak oluşturulmuştur. Üretim verileri, EPİAŞ Şeffaflık Platformu’ndan alınmış ve Python ile işlenmiştir. XGBoost modeli, farklı ağaç sayıları ve öğrenme oranı (η) denenerek optimize edilmiştir. Modelin etkinliği, belirleme katsayısı (R²), Ortalama Mutlak Ölçekli Hata (MASE), Kök Ortalama Karesel Hata (RMSE), Ortalama Mutlak Hata (MAE) ve Ağırlıklı Mutlak Yüzdesel Hata (WAPE) gibi çeşitli hata ölçümleri ile titizlikle değerlendirilmiştir. Bu çalışmada kullanılan yöntemler ve elde edilen sonuçlar, hidroelektrik enerji tahmininde makine öğrenimi algoritmalarının faydalı olabileceğini ve enerji yönetimi stratejilerinin optimize edilmesine yönelik önemli bilgiler sunabileceğini göstermektedir.

Anahtar Kelimeler

• HES
• Hidroelektrik
• Tahmin
• XGBoost

Kaynakça

1. 1. T.C. Enerji ve Tabii Kaynaklar Bakanlığı, Elektrik Bilgi Merkezi, https://enerji.gov.tr/bilgi-merkezi-enerji-elektrik, Erişim tarihi: 23.09.2024.
2. 2. Acaroğlu, H., Kartal, H. M. & García Márquez, F. P. (2023). Testing the environmental Kuznets curve hypothesis in terms of ecological footprint and CO2 emissions through energy diversification for Turkey. Environmental Science and Pollution Research, 30(22), 63289-63304.
3. 3. Cassagnole, M., Ramos, M.-H., Zalachori, I., Thirel, G., Garçon, R., Gailhard, J. & Ouillon, T. (2021). Impact of the quality of hydrological forecasts on the management and revenue of hydroelectric reservoirs – a conceptual approach. Hydrology and Earth System Sciences, 25(2), 1033-1052.
4. 4. IEA, Energy Statistics Data Browser, https://www.iea.org/data-and-statistics/data-tools/energy-statistics-data-browser, Erişim tarihi: 20.09.2024.
5. 5. EÜAŞ Aslantaş HES. https://www.euas.gov.tr/santraller/aslantas-hes. Erişim tarihi: 20.09.2024.
6. 6. Hamel, P., Bremer, L.L., Ponette-González, A.G., Acosta, E., Fisher, J.R.B., Steele, B., Cavassani, A. T., Blainski, E. & Brauman, K.A. (2020). The value of hydrologic information for watershed management programs: The case of Camboriú, Brazil. Science of The Total Environment, 705, 135871.
7. 7. Ercüment Beyhun, N., Altintaş, K.H. & Noji, E. (2005). Analysis of registered floods in Turkey. International Journal of Disaster Medicine, 3(1-4), 50-54.
8. 8. O’Connor, R.E., Yarnal, B., Dow, K., Jocoy, C.L. & Carbone, G.J. (2005). Feeling at risk matters: water managers and the decision to use forecasts. Risk Analysis, 25(5), 1265-1275.

9. 9. Özesmi, U. & Özesmi, S. (2003). A participatory approach to ecosystem conservation: fuzzy cognitive maps and stakeholder group analysis in uluabat lake, Turkey. Environmental Management, 31(4), 518-531
10. 10. Chen, T. & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785-794. New York, NY, USA: ACM.
11. 11. Anghileri, D., Monhart, S., Zhou, C., Bogner, K., Castelletti, A., Burlando, P. & Zappa, M. (2019). The Value of Subseasonal Hydrometeorological Forecasts to Hydropower Operations: How Much Does Preprocessing Matter? Water Resources Research, 55(12), 10159-10178.
12. 12. Kumar, V., Kedam, N., Sharma, K.V., Mehta, D.J. & Caloiero, T. (2023). Advanced machine learning techniques to improve hydrological prediction: a comparative analysis of streamflow prediction models. Water, 15(14), 2572.
13. 13. Szczepanek, R. (2022). Daily streamflow forecasting in mountainous catchment using XGBoost, LightGBM and CatBoost. Hydrology, 9(12), 226.
14. 14. Tayfur, G., Singh, V., Moramarco, T. & Barbetta, S. (2018). Flood hydrograph prediction using machine learning methods. Water, 10(8), 968.
15. 15. Hao, R. & Bai, Z. (2023). Comparative study for daily streamflow simulation with different machine learning methods. Water, 15(6), 1179.
16. 16. Zhang, D., Qian, L., Mao, B., Huang, C., Huang, B. & Si, Y. (2018). A Data-driven design for fault detection of wind turbines using random forests and XGBoost. IEEE Access, 6, 21020-21031.
17. 17. Liu, Y., Luo, H., Zhao, B., Zhao, X. & Han, Z. (2018). Short-term power load forecasting based on clustering and XGBoost method. 2018 IEEE 9th International Conference on Software Engineering and Service Science (ICSESS), Beijing, China.
18. 18. Zheng, H. & Wu, Y. (2019). A XGBoost model with weather similarity analysis and feature engineering for short-term wind power forecasting. Applied Sciences, 9(15), 3019.
19. 19. Abbasi, R.A., Javaid, N., Ghuman, M.N.J., Khan, Z.A., Ur Rehman, S. & Amanullah (2019). Short term load forecasting using XGBoost, 1120-1131.
20. 20. Suo, G., Song, L., Dou, Y. & Cui, Z. (2019). Multi-dimensional short-term load forecasting based on XGBoost and fireworks algorithm. 2019 18th International Symposium on Distributed Computing and Applications for Business Engineering and Science (DCABES), Wuhan, China.
21. 21. Li, C., Chen, Z., Liu, J., Li, D., Gao, X., Di, F., Li, L. & Ji, X. (2019). Power load forecasting based on the combined model of LSTM and XGBoost. PRAI '19: Proceedings of the 2019 the International Conference on Pattern Recognition and Artificial Intelligence, New York, NY, USA: ACM.
22. 22. Liao, X., Cao, N., Li, M. & Kang, X. (2019). Research on short-term load forecasting using XGBoost based on similar days. 2019 International Conference on Intelligent Transportation, Big Data & Smart City (ICITBS), 675-678, Changsha, China.
23. 23. Guo, X., Gao, Y., Zheng, D., Ning, Y. & Zhao, Q. (2020). Study on short-term photovoltaic power prediction model based on the Stacking ensemble learning. Energy Reports.
24. 24. Phan, Q.-T., Wu, Y.-K. & Phan, Q.-D. (2020). A comparative analysis of XGBoost and temporal convolutional network models for wind power forecasting.
25. 25. Al Rayess, H.M. & Ülke Keskin, A. (2021). Forecasting the hydroelectric power generation of GCMs using machine learning techniques and deep learning (Almus Dam, Turkey). Geofizika, 38(1), 1-14.
26. 26. Zhang, K., Gu, C., Gu, C., Zhu, Y., Chen, S., Dai, B. & Li, Y. (2021). A novel seepage behavior prediction and lag process identification method for concrete dams using HGWO-XGBoost model. IEEE Access.
27. 27. Ma, M., Zhao, G., He, B., Li, Q., Dong, H., Wang, S. & Wang, Z. (2021). XGBoost-based method for flash flood risk assessment. Journal of Hydrology, 598, 126382.
28. 28. Osman, A.I.A., Ahmed, A.N., Chow, M.F., Huang, Y.F. & El-Shafie, A. (2021). Extreme gradient boosting (XGBoost) model to predict the groundwater levels in Selangor Malaysia. Ain Shams Engineering Journal.
29. 29. Zhang, W., Wei, Z., Wang, H., Hanyong, W., Lin, Y., Yemin, L., Liu, W. & An, X. (2021). Reservoir inflow predicting model based on machine learning algorithm via multi-model fusion: A case study of Jinshuitan river basin. IET Cyber-Systems and Robotics.
30. 30. Bae, D.-J., Kwon, B.-S. & Song, K.-B. (2021). XGBoost-based day-ahead load forecasting algorithm considering behind-the-meter solar PV generation. Energies, 15(1), 128.
31. 31. Wang, Y., Sun, S., Chen, X., Zeng, X., Kong, Y., Chen, J., Guo, Y. & Wang, T. (2021). Short-term load forecasting of industrial customers based on SVMD and XGBoost. International Journal of Electrical Power & Energy Systems.
32. 32. Phan, Q.-T., Wu, Y.-K. & Phan, Q.-D. (2021). Short-term solar power forecasting using XGBoost with numerical weather prediction. IEEE International Future Energy Electronics Conference, Taipei, Taiwan.
33. 33. Phan, Q.T., Wu, Y.K. & Phan, Q.D. (2021). A hybrid wind power forecasting model with XGBoost, data preprocessing considering different NWPs. Applied Sciences, 11(3), 1100.
34. 34. Wang, B., Li, T., Xu, N., Zhou, H., Xiong, Z. & Long, W. (2021). A novel reservoir modeling method based on improved hierarchical XGBoost. 2021 IEEE 5th Advanced Information Technology, Electronic and Automation Control Conference (IAEAC), Chongqing, China.
35. 35. Udo, W. & Muhammad, Y. (2021). Data-driven predictive maintenance of wind turbine based on SCADA data. IEEE Access, 9, 162370-162388.
36. 36. Chen, Z., Xiao, J., Chen, S., Qiao, H., Chen, J. & Xu, X. (2022). Shaft run-out trend prediction of water turbine generators and fault identification of hydroelectric units based on XGBoost algorithm. International Conference on Measuring Technology and Mechatronics Automation, Changsha, China.
37. 37. Fan, L., Wang, Y., Fang, X. & Jiang, J. (2022). To predict the power generation based on machine learning method. Journal of Physics: Conference Series, 2310(1), 012084.
38. 38. Singh, U. & Rizwan, M. (2022). SCADA system dataset exploration and machine learning based forecast for wind turbines. Results in Engineering, 16, 100640.
39. 39. Xue, J., Hu, X., Chen, H. & Zhou, G. (2022). Research on LSTM-XGBoost integrated model of photovoltaic power forecasting system.
40. 40. Mohamed, M., Mahmood, F.E., Abd, M.A., Chandra, A. & Singh, B. (2022). Dynamic forecasting of solar energy microgrid systems using feature engineering. IEEE Transactions on Industry Applications, 58(6), 7857-7869.
41. 41. Xiong, X., Guo, X., Zeng, P., Zou, R. & Wang, X. (2022). A short-term wind power forecast method via XGBoost hyper-parameters optimization. Frontiers in Energy Research, 10.
42. 42. Wang, J., Gao, Z. & Ma, Y. (2022). Prediction model of hydropower generation and its economic benefits based on EEMD-ADAM-GRU fusion model. Water, 14(23), 3896.
43. 43. Hong, Y., Yang, J., Yang, Z. & Yan, J. (2023). PV power prediction based on XGBoost algorithm. 2023 IEEE 5th International Conference on Civil Aviation Safety and Information Technology (ICCASIT), 930-933. IEEE.
44. 44. Zou, Y., Chen, H., Zhang, Y., Wang, Z., Pang, L., Lan, X., Yitong, Z., Wang, B. & Peng, R. (2023). Multimode hydropower power prediction based on long short-term memory. Sixth International Conference on Computer Information Science and Application Technology (CISAT 2023), Hangzhou, China.
45. 45. Tran, N.T., Tran, T.T.G., Nguyen, T.A. & Lam, M.B. (2023). A new grid search algorithm based on XGBoost model for load forecasting. Bulletin of Electrical Engineering and Informatics, 12(4), 1857-1866.
46. 46. Sun, S., Xing, J., Cheng, Y., Yu, P., Wang, Y., Yang, S. & Wang, S. (2023). Features analysis and prediction of electric load based on clustering and XGBoost. Second International Conference on Energy, Power, and Electrical Technology (ICEPET 2023), Kuala Lumpur, Malaysia.
47. 47. Wu, Y., Xie, Y., Xu, F., Zhu, X. & Liu, S. (2024). A runoff-based hydroelectricity prediction method based on meteorological similar days and XGBoost model. Frontiers in Energy Research, 11.
48. 48. Bashir, S.B., Farag, M.M., Hamid, A.K., Adam, A.A., Abo-Khalil, A.G. & Bansal, R. (2024). A novel hybrid CNN-XGBoost model for photovoltaic system power forecasting. 2024 6th International Youth Conference on Radio Electronics, Electrical and Power Engineering, Moscow, Russian Federation.
49. 49. EPİAŞ, Şeffaflık Platformu Gerçek Zamanlı Elektrik Üretim Verisi, https://seffaflik.epias.com.tr/ electricity/electricity-generation/ex-post-generation/real-time-generation, Erişim tarihi: 23.09.2024.
50. 50. Global Modeling and Assimilation Office (GMAO) (2015). MERRA-2 inst1_2d_asm_Nx: 2d, 1-Hourly, Instantaneous, Single-Level, Assimilation, Single-Level Diagnostics V5.12.4. Greenbelt, MD, USA: Goddard Earth Sciences Data and Information Services Center (GES DISC). Erişim tarihi: 23.09.2024.
51. 51. Global Modeling and Assimilation Office (GMAO) (2015). MERRA-2 tavg1_2d_rad_Nx: 2d, 1-Hourly, Time-Averaged, Single-Level, Assimilation, Radiation Diagnostics V5.12.4. Greenbelt, MD, USA: Goddard Earth Sciences Data and Information Services Center (GES DISC). Erişim tarihi: 23.09.2024.
52. 52. Aiken, E. (2018). Script to download data from MERRA-2 database based on lat/lon coordinates. Github. https://github.com/emilylaiken/merradownload, Erişim tarihi: 01.03.2024.
53. 53. DMLC, eXtreme Gradient Boosting. Distributed (Deep) Machine Learning Community, https://github.com/dmlc/xgboost, Erişim tarihi: 28.05.2024.
54. 54. Morde, V. & Setty, A., XGBoost Algorithm: Long May She Reign! Towards Data Science. https://towardsdatascience.com/https-medium-com-vishalmorde-xgboost-algorithm-long-she-may-rein- edd9f99be63d, Erişim tarihi: 23.09.2024.