Estimation of wind power density with artificial neural network

Industry and technology are rapidly developing with each passing day. They need energy to sustain this evolution. The demand of energy is mainly provided from fossil fuels. Unfortunately, this kind of energy reserves are consumed away day by day. Therefore, there is a need to use alternative energy sources to supply energy needs. Alternative energy sources can be listed as; solar, wind, wave, biomass, geothermal and hydro-electric power. Our country has significant potential for wind energy. Wind power density estimation is required to determine the wind potential. In this study, the wind power density was estimated by using artificial neural network (ANN) method. Forty meteorological station were used for ANN training, while eighteen meteorological stations were used to test the trained network. Network has trained according to, respectively; trainlm, trainbfg, trainscg, traincgp traincgb, traincgf ve trainoss learning algorithms. The correlation coefficient (R) and Mean bias error (MBE) of the best developed model were calculated as 0,9767 and -0,3124 W/m^2 respectively. Root Mean Square Error (RMSE) was calculated as 1,4786 W/m^2. In conclusion, the obtained results demonstrate that the developed model can be used to estimate the wind power density. Keywords: Alternative energy sources, wind power density, wind speed, artificial neural networks