Abstract: The variation characteristics of input wind speed directly affect the output power modeling of wind turbine. In this paper, nonlinear least square method and differential evolution algorithm are used to calculate parameters, and Logistic function is used to model the output power of wind turbine, combined with the high accuracy of heuristic algorithms in finding the optimal construction. Then, the modeling accuracy of the three methods for output power model is compared. The results show that the nonlinear least squares method is more suitable for the calculation of the proposed model parameters. The high accuracy of the traditional output power model is derived from the artificially defined rated power value in the model, while the Logistic function output power model does not need to artificially define the rated power value, the modeling accuracy of which is better under the same conditions.

Key words: wind turbine, power characteristic model, Logistic function