Dynamic Equivalent Modelling of Wind Farm Frequency Regulation Based on Koopman Theory
A dynamic equivalent modelling method based on Koopman theory was proposed to address the issues of low accuracy and high model complexity in equivalent modelling under dynamic operating conditions of wind farms.Firstly,based on the actual operating data of the wind farm,the dynamic process curves of each wind turbine participating in frequency regulation were obtained,and the data sampling and processing were completed;secondly,based on the Koopman theory,starting from the observable measurement of wind turbines,the evolution of the correlation function defined in the state space was investigated,and the nonlinear wind turbine system was approximately transformed into a linear system in a higher dimensional function space,obtaining the κ operator that can describe the frequency modulation dynamic characteristics of each wind turbine;then,the gap measurement method was used to cluster and group the κ operators of each unit,and an equivalent model of the wind farm was established based on the clustering results;finally,the accuracy of the proposed method was demonstrated through numerical simulations.The proposed method can effectively characterize the complex response characteristics of wind farms participating in frequency regulation,improving the accuracy of the equivalent model.
wind farmprimary frequency controlKoopman theorygap measurementequivalent modelling