Aggregate Equivalent Modeling of Wind Farm under Full Wind Speed Scenarios

Currently, real-time collection of operational data at the individual wind farm unit level for equivalent modeling is not feasible within the existing power grid infrastructure. Hence, this paper introduces a methodology for wind farm aggregation equivalent modeling under full wind speed scenarios. Initially, a year's worth of operational data from wind farm units is gathered, and the Fuzzy C-means (FCM) clustering technique is applied. This facilitates the extraction of typical wind farm operational scenarios by clustering unit wind speed data over the entire year. Subsequently, a four-machine clustering approach is employed to categorize these typical operational scenarios, resulting in various unit groupings. Following this, distinct multi-machine equivalent models are developed for different scenarios arising from diverse group combinations. A typical fault simulation is conducted to determine the equivalent error associated with various unit groupings, selecting the group combination with the minimal error as the optimal clustering strategy. Lastly, by integrating the parameters calculation for equivalent wind turbines and collector networks, the aggregate equivalent modeling of the wind farm is formed. The model's effectiveness is validated using data from an actual wind farm.