The comprehensive load model with wind power generation involved is established based on the load model of distribution network. The overall measuring identification method is then used for parameter identification with measured node voltages. Substituting the node voltage into mathematical model of load, the corresponding current can be calculated through solving its state equation with improved Euler method. The weighted error between measured and calculated currents is selected as the objective function, which can be solved with the genetic algorithm to determine the parameters of the load model. Given measured PMU data from north China power grid, parameters of comprehensive load model are identified with the method proposed above considering wind power generation. The identified results reflect that the main part of load are double fed induction generators for wind power generation, which is in good agreement with the characteristic of load and verifies both the validity of identification method and rationality of identification results.