Abstract:Modeling on the dynamic behavior of wind farm accurately is the premise of studying the impact of large-scale wind power integrated into power system. Considering the difference problem of wind speed because of complex terrain and wind turbine units arranged irregularly in large wind farm, long-time scale measured data is chosen to be the cluster-dependent index. The wind turbine generators are divided into groups by the method of improved fuzzy Cmeans clustering algorithm. Furthermore, parameter identification for equivalent machine model is provided based on the particle swarm optimization algorithm of global optimum location mutation. The simulation results show that dynamic equivalent model of the wind farm established can reflect the dynamic nature at the point of common coupling (PCC) accurately. It can be used to analyze the stability of large-scale wind power with double fed induction generator (DFIG) integrated into power system, which has important value in engineering.

Dynamic Equivalent Modeling of Wind Farm with Double Fed Induction Wind Generator Based on Operating Data