Energy production estimation in hyroelectric power plants using machine learning methods: An application in Erzincan

Electricity, a secondary energy source that is considered a strategic concept for the continuity and development of countries social and economic life cycles, is undoubtedly the most widely used type of energy. Knowing electricity production through studies using various methods will assist create advantages for the sustainability of energy in the future. In this thesis, it is aimed to create the most appropriate model for short and medium term production estimates in hydroelectric power plants (HPP) by using machine learning methods. It is observed that small capacity, which is mostly channel type, HEPPs have been established with increasing demand for renewable energy sources in electricity production in the recent years. It has become important to create forecast models that can provide reliable results and work with high accuracy in order to determine the future production amount of existing and planned HEPPs of channel type, to see the instantaneous production power and to make income evaluation in terms of investment. It is thought that the study will be a case study for production planning in channel type HEPPs and estimating the amount of energy production in regions where new investments will be made.