Dynamic Equivalent Modeling of Wind Farm Based on IGWO-K-means Method

The equivalent modeling of wind farm is the premise and basis for the analysis of wind power system. In order to improve the precision of dynamic equivalence of wind farm and reduce the difficulty of equivalence, 14 variables from the running data of the 24 wind turbines in an Inner Mongolia wind farm are selected as clustering index based on such four aspects as wind(wind speed and wind direction), induction machine, wind power output effect and working environment, which comprehensively describes the characteristics of the wind farm. Secondly, two control strategies, nonlinear strategy for convergent factor and dynamic reference rate strategy, are introduced to improve Grey Wolf Optimizer algorithm (GWO). In addition, the best cluster centers are searched by combining with K-means clustering algorithm, the clustering results are outputted, and a dynamic equivalent model of wind farm is built. Finally, a clustering model is established on MATLAB/Simulink platform to verify the feasibility of the model. The results show that the method can improve the accuracy of wind farm equivalent modeling and describe the dynamic characteristics of wind farms more accurately.