Evaluating Ai-Based Models for Energy Economy Nexus Analysis: a Focus on Hydroelectric Power

The global transition to renewable energy highlights the importance of forecasting hydroelectric power generation, which is influenced by economic and demographic factors. This study examines how key indicators, such as GDP, energy consumption, population, and inflation, relate to hydroelectric power generation using Artificial Neural Networks (ANNs), using a dataset spanning 42 years (19802021). Three ANN models were developed: one using GDP and energy consumption, another incorporating population, and a third replacing population with inflation. The results suggest that including population improves the model's predictive ability, with an R-value of 0.9969 and MSE of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$6.4307 \mathrm{e}+03$</tex>. At the same time, inflation introduces variability, leading to a higher test MSE of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$1.8151 \mathrm{e}+04$</tex> and reduced robustness. These findings highlight the influence of demographic factors on forecasting performance and suggest that selecting appropriate predictors can support energy planning and decision-making.