Yapay sinir ağları ile YEKDEM birim fiyat tahminlemesi

With the developing technology, energy production methods are changing in the world. Fossil energy resources have begun to be abandoned and replaced by renewable energy resources with unlimited capacity. In this way, both green energy is produced and less cost is spent. Improvements are also available for the production of green energy in Turkey. With the YEK Law enacted in 2005, YEKDEM (Renewable energy resources support mechanism) started. YEKDEM is a support mechanism applied to wind, biomass, canal, solar, gas obtained from biomass, wave current energy, geothermal, tidal energy, river and hydroelectric production facilities. Energy produced by the Yekdem method also has a monthly updated unit price. This unit price is named as yekdem unit price. In this present studying, an unit price prediction has been made with ann, which are one of the sub-branches of artificial intelligence. In the project, four dependent variables were estimated and the unit price of the same was obtained with these variables. In the estimation process, 30 independent parameters were used. Different machine learning models were created to compare the results of neural networks and the best result was investigated. MPE and MSE were used as accuracy criteria in the study. It was 0.35 in the MSE criterion in the best model and 53.67 in the MSE criterion in the worst model.