Wind energy is a renewable source of energy which is clean, freely available, unlimited, and freely distributed throughout the world. Accurate, wind turbine power curve models needed to estimate annual energy production of wind farms and monitoring the performance of wind turbines. This study conducted at Adama wind farm II with the aim to compare four wind turbines power curve models for estimation of annual energy production based on 5mindata form the years of 2016 to 2020. Using MS Excel, MATLAB software and statistical model the wind data has been analyzed to estimate power density, annual energy production (AEP), Capacity factor, mean wind speed mean wind power, Weibull shape parameter (K) and scale parameter(C). Analyzing a wind farm data at a height of 70m using MS Excel and MATLAB software mean wind speed of 7.4329m/s, mean wind power density 479W/m 2 , Weibull shape parameter 3.23, Weibull scale parameter 8.38m/s, and Capacity factor 35.3% have been found at the farm.The data collected from nineteen sample wind turbines selected as a sample size from 102 wind turbines using simple random sampling techniques. Two turbines N46 and S24 selected to compare four wind turbine power curve models namely cubic type two model(CTTM), general model(GM), exponential model(EM) and approximate power coefficient model. These power curve models compared using mean absolute percentage error (MAPE) and root mean square error (RMSE).Among the four-wind turbine power curve models exponential model was the best model. Key words: Annual energy production, Capacity factor, Scale factor, Shape factor, Weibull distribution, Wind turbines power curve models