Aiming at the problem of low accuracy of wind farm equivalent models under fault scenarios

a multi-machine aggregation identification equivalence modeling method for wind farm considering low voltage ride through (LVRT) power characteristics is proposed. Firstly

the wind farm is initially grouped based on the typical differences in active power dynamic characteristics during LVRT of wind turbines. Secondly

the Ng-Jordan-Weiss (NJW) algorithm based on dynamic time warping (DTW) metric is used to realize the secondary division of the cluster

and the final two-stage clustering results are obtained. Then

for the aggregated multi-machine equivalent model of the wind farm

parameter sensitivity analysis is used to determine the key parameters to be optimized. The aggregated values of each parameter are taken as the initial values

and the overall equivalent parameters of the wind farm are optimized using the three strategies of single-machine step-by-step identification

multi-machine sequential identification

and equivalent impedance identification. Finally

the fitting curves and equivalent errors of different methods are compared

and the results show that the proposed method effectively improves the accuracy and adaptability of the equivalent model.