Modeling the Impact of Climate Change on Hydroelectric Production Using Machine Learning Approaches: The Case of Menzelet Dam in the Ceyhan Basin

Authors : Ezgi Öztürk İspir, Şerife Yurdagül Kumcu, H. Erdinç Koçer

Pages : 131-144

Doi:10.59940/jismar.1719556

View : 46 | Download : 146

Publication Date : 2025-12-31

Article Type : Research Paper

Abstract :The main objective of this study is to model the effects of climatic and hydrological variability on hydroelectric energy production and to provide a scientific basis for future water-energy management scenarios. At the same time, to test the applicability of machine learning approaches in energy estimation models and to provide a guiding analysis for decision makers at the regional scale. For this purpose, hydroelectric energy production estimation was performed on daily hydro-meteorological observation data (precipitation, temperature, evaporation) for the years 1999–2020 using machine learning model approach. Random Forest, Decision Tree, Gradient Boosting and Support Vector Regression algorithms were used for estimation in the modeling process; each model was subjected to hyperparameter optimization with the GridSearchCV method. Model success was compared with error metrics such as R², MAE and MSE and the highest success algorithm was determined. Using the best approach determined, future energy production estimates were carried out with the projection data obtained from HadGEM2-ES, GFDL-ESM2M and MPI-ESM-MR climate models for the period 2023–2098 (under RCP 4.5 and RCP 8.5 scenarios). The study will make a significant contribution to literature in terms of evaluating the effects of climate change on hydroelectric energy potential.
Keywords : Enerji tahmini, İklim modelleri, Makine öğrenmesi

ORIGINAL ARTICLE URL