Trend Curve- and Machine Learning-Based Renewable Energy Development Forecast

While the energy demand is rapidly increasing, allocating enough sources becomes a major task. And since today's outlook is shaped by a united economical-technical-environmental aspect, identifying the renewable energy (RE) place is vital. Therefore, this paper presents the trend of leading non-hydro RE sources using regression based on machine learning, s-shaped curves, i.e., logistic and gompertz, and a hybrid weighted method until 2035. Results show that regardless of the rate or beginning time, a saturation in RE trend is expected. This saturation will happen even after RE well-known trend in recent years. Further, the RE installed capacity in its leading countries until 2035 is forecasted. And based on the results, a discussion is provided to show how to postpone this saturation.