With the capacity of installed wind power generators steadily increasing in China， power fluctuation from wind farms will significantly affect the security and reliability of the power system. Traditional modeling of power fluctuation is based on the time domain or statistics methodology which， however， is incapable of completely explaining the physical mechanism of randomness of power fluctuation. To remedy such a situation， fluctuation modeling based on the frequency domain is proposed. The frequency domain characteristics of stochastic fluctuation on large wind farms are studied using the power spectral density of wind speed， the frequency domain model of a wind power generator and the information on weather and geography of the wind farms. The correctness and effectiveness of the model are verified by comparing the measurement data with simulation results of a certain wind farm.